US Federal Reserve: 121702CFA NCRA Credit Score Report Final

8/14/2019 US Federal Reserve: 121702CFA NCRA Credit Score Report Final

http://slidepdf.com/reader/full/us-federal-reserve-121702cfa-ncra-credit-score-report-final 1/51

Credit Score Accuracy and Implications for Consumers

December 17, 2002

Consumer Federation of America

National Credit Reporting Association



ii

Table of Contents

I. About Privacy ......................................................................................................... 1II. The Growing Importance of Credit Scores............................................................... 2III. Controversial Issues Affecting Consumers........................................................... 4

A. Speed...................................................................................................................4B. Customized or Risk-Based Pricing....................................................................... 4C. Effect on Discrimination...................................................................................... 4D. Statistical Validity ............................................................................................... 5E. Untested Scoring Formulas.................................................................................. 5F. Inaccurate credit reports....................................................................................... 6

IV. How Does the System Work? .............................................................................. 8A. Non-Mortgage Credit .......................................................................................... 9B. Employment and Services Other Than Loans.....................................................10C. Other Data Providers ......................................................................................... 10D. Mortgage Credit ................................................................................................ 11

1. Portfolio Loans.............................................................................................. 112. Loans Sold in the Secondary Market.............................................................. 133. Credit Rescoring............................................................................................ 144. Federal Housing Administration (FHA) and Department of Veterans’ Affairs(VA) Loans ........................................................................................................... 15

V. Study Design......................................................................................................... 16A. Phase One.......................................................................................................... 16B. Phase Two......................................................................................................... 17C. Phase Three....................................................................................................... 18

VI. Findings ............................................................................................................ 20A. Phase One.......................................................................................................... 20

1. Almost One in Ten Files was Missing a Credit Score from at Least OneRepository............................................................................................................. 202. A Substantial Number of Files Met the Criteria for Further Review. .............. 203. Numerous Files Contained Additional Repository Reports and Information notRelevant to the Consumer’s Credit History............................................................ 214. Scores Reported by the Three Repositories for a Given Consumer VariedSubstantially.......................................................................................................... 225. Reports Contained Limited Information to Help Consumers Understand thePrincipal Reasons for their Credit Scores............................................................... 236. In Depth Reviews Revealed Significant Errors and Inconsistencies, Some of Which were Likely Artificially Lowering Consumer Credit Scores, and Some of Which were Likely Artificially Raising Consumer Credit Scores. .......................... 24

B. Phase Two......................................................................................................... 251. Scores Reported by the Three Repositories for a Given Consumer VariedSubstantially.......................................................................................................... 252. Reports Scored With Different Versions of Scoring Software Reflected AlmostNo Difference in Overall Variability of Credit Scores............................................ 263. Reports Contained Limited Information to Help Consumers Understand thePrincipal Reasons for their Credit Scores............................................................... 27



iii

C. Phase Three – Specific Types of Errors ............................................................. 281. Significance and Frequency of Errors of Omission......................................... 302. Errors of Commission .................................................................................... 323. Merging and Compilation Errors.................................................................... 34

VII. Conclusions and Implications of the Findings for Consumers ............................ 37

A. Credit scores and the information in credit reports vary significantly amongrepositories................................................................................................................ 37B. Many consumers are unharmed by these variations, and some probably benefitfrom them.................................................................................................................. 37C. However, tens of millions of consumers are at risk of being penalized forincorrect information in their credit report, in the form of increased costs or decreasedaccess to credit and vital services............................................................................... 37D. Almost one in ten consumers runs the risk of being excluded from the creditmarketplace altogether because of incomplete records, duplicate reports, and mixedfiles. 39E. The use of information from all three repositories in mortgage lending protects

consumers and creditors from being negatively affected by errors of omission, but itmay increase the negative impact on consumers of errors of commission. ................. 40F. Consumers are not given useful and timely information about their credit.......... 41

1. Standardized, generic explanations do not provide sufficient information forconsumers to address inconsistencies and contradictions, let alone outright errors. 412. Consumers outside of California have no affirmative right to know their creditscores. ................................................................................................................... 41

G. Private companies without significant oversight are setting, or at the very leastheavily influencing, the rules of the marketplace for essential consumer services thatbase decisions on credit scores. ................................................................................. 42H. Certain information in credit reports has the potential to cause breaches of consumers’ medical privacy. ..................................................................................... 42

VIII. How to Improve the System...............................................................................44A. Require creditors to immediately provide to any consumer who experiences anadverse action as a result of their credit reports or credit scores a copy of the creditreports and scores used to arrive at that decision free of charge and permit disputes tobe immediately resubmitted for reconsideration.........................................................44B. Require decisions based on a single repository’s credit report or credit score thatresult in anything less than the most favorable pricing to immediately trigger a re-evaluation based on all three repositories at no additional cost................................... 44C. Strengthen requirements for complete and accurate reporting of accountinformation to credit repositories, and maintenance of consumer data by therepositories, with adequate oversight and penalties for non-compliance..................... 45D. Establish meaningful oversight of the development of credit scoring systems. ... 46E. Address important questions and conduct further research. ................................ 47

IX. Recommendations for Consumers...................................................................... 48



1

I. About Privacy

The Consumer Federation of America (CFA) and the National Credit ReportingAssociation (NCRA) designed the details of this study with advice from legal counsel toensure the methodology would comply with the requirements of the Fair Credit Reporting

Act, Gramm Leach Bliley Act, and other consumer privacy laws. From the outset, eachorganization was mindful of the ethical spirit and intent of these consumer protection andprivacy laws. In this day of rampant identification theft, we carefully evaluated eachsegment of the study workflow to ensure that we analyzed data extracted from the creditfiles without any trace of personal identifiers. Regarding consumer identity, all non-public, personal information data was completely “blind” as to a source for analysis. Nonames, addresses, social security numbers, dates of birth, account numbers, or any otheritem that could be used in any way to trace back to a specific consumer were revealed toor recorded by any third party outside trusted personnel of the consumer reportingagencies involved in the study. In one phase of the study the recorded data segmentclosest to the consumer was the postal zip code of their residence.

After CFA made a random selection of the time frame from which credit files were to beanalyzed, a generic number was assigned to keep the nameless study data from eachstudy file separated from other study files. No copies or partial copies of any creditreports, on paper or electronically, were removed from any credit reporting agencylocation. Anonymous credit scores and an analysis of the credit data, as reviewed bycredit reporting agency personnel for security and industry knowledge, was supervisedand recorded by the CFA researcher for tabulation. The data elements recorded in thisstudy are insufficient to ever be used to track or identify any individual. Further, theanalytical data recorded, if ever obtained by unscrupulous individuals, contains noinformation that could ever be used to try to defraud any of the consumers or creditorsconnected to the files in the study. Total anonymity to consumer identity and creditoraccounts was, and will continue to be, strictly enforced.



2

II. The Growing Importance of Credit Scores

Consumer access to credit, housing, insurance, basic utility services, and evenemployment is increasingly determined by centralized records of credit history andautomated interpretations of those records.

Credit histories in one form or another have long been an important factor in decisions toextend or deny credit to consumers1. Historically, such decisions required a skilled,human evaluation of the information in an applicant’s credit history to determine thelikelihood that the applicant would repay a future loan in a timely manner. Morerecently, computer models have been developed to perform such evaluations. Thesemodels produce numerical credit scores that function as a shorthand version of anapplicant’s credit history to facilitate quick credit assessments.

During the second half of the 1990s, mortgage underwriting increasingly incorporatedcredit scores and other automated evaluations of credit histories. As of 1999,

approximately 60 to 70 percent of all mortgages were underwritten using an automatedevaluation of credit, and the share was rising2.

The automated quantification of the information in credit reports has not simply beenused to decide whether or not to extend credit, but has also been used to set prices andterms for mortgages and other consumer credit. In certain cases, even very smalldifferences in scores can result in substantially higher interest rates, and less favorableloan terms on new loans. Credit scores are also used to determine the cost of privatemortgage insurance, which protects the lender, not the consumer, from loss but isrequired on mortgages with down payments of less than twenty percent3. Lenders alsoreview credit histories and/or credit scores to evaluate existing credit accounts, and usethe information when deciding to change credit limits, interest rates, or other terms onthose accounts.

In addition to lenders, potential landlords and employers may review credit historiesand/or credit scores. Landlords may do so to determine if potential tenants are likely topay their rent in a timely manner. Employers may review this information during a hiringprocess, especially for positions where employees are responsible for handling large sumsof money. Utility providers, home telephone, and cell phone service providers also mayrequest a credit report or credit score to decide whether or not to offer service toconsumers.

Insurance companies have also begun using credit scores and similar insurance scores –that are derived from the same credit histories – when underwriting consumerapplications for new insurance and renewals of existing policies. Credit information has

1 Klein, Daniel. 2001. Credit Information Reporting. Why Free Speech is Vital to Social Accountabililyand Consumer Opportunity. The Independent Review. Volume V, number 3.2 Straka, John. 2000. A Shift in the Mortgage Landscape: the 1990s Move to Automated CreditEvaluations. Journal of Housing Research. Volume 11, Issue 2.3 Harney, Ken. August 18, 2002. “Risk-based pricing brings a big rate hike for some.” Washington Post.



3

been used as a basis to raise premiums, deny coverage for new customers, and denyrenewals of existing customers – even in the absence of other risk factors, such as movingviolations or accidents. Some providers claim that credit scores are also used to offerinsurance coverage to consumers who have previously been denied, or to lower insurancerates. This is a highly contested issue that is under review in dozens of state legislatures

and insurance commissions.

Thus, a consumer’s credit record and corresponding credit score can determine accessand pricing for the most fundamental financial and consumer services.



4

III. Controversial Issues Affecting Consumers

The expanded use of automated credit evaluations has brought changes to themarketplace that have benefited consumers. However, given the tremendous impactcredit scores can have on consumers’ ability to access and afford basic necessities, the

increased application of this tool has also raised serious concerns about the potentialharm it can cause.

A. Speed

The growth in use of credit scores has dramatically increased the speed at which manycredit decisions can be made. Especially for consumers with relatively good credit,approvals for loans can be given in a fraction of the time previously required, without anymanual review of the information. It is unlikely that underwriting the recent recordvolumes of mortgage originations would have been possible without the efficienciesprovided by credit scoring.

B. Customized or Risk-Based Pricing

Credit scores, as a quantitative shorthand for credit histories, increase the potential forcustomized pricing of credit based on the risk an individual poses. Some argue thatcharging more to consumers defined as higher risk would remove some of the cost of risk carried by the general consumer population, and would allow for price reductions amongconsumers who pose less risk. Others argue that the savings have not been – and areunlikely to be – passed on to consumers who pose less risk, and scoring systems simplyallow lenders to extract greater profits from consumers who do not attain target creditscores. The potential for increased profits from consumers whose credit is scored lowalso creates a disincentive to helping consumers correct errors in their credit records.

The increased speed at which underwriting decisions can be made has created pressure tocomplete credit applications more quickly. Some contend that the combination of thisincreased pace and the increased ability to customize the price charged based on creditallows lenders to approve a larger share of consumers for loans, but not necessarily at thebest rates for which they qualify. While many consumers can feel overwhelmed by largecredit based transactions, such as mortgage closings, consumers who do not have a solidunderstanding of credit scores, or who do not objectively know their creditworthiness, areeven more vulnerable to high-pressure tactics to accept any offer of credit, regardless of terms, and may unnecessarily be charged higher rates.

C. Effect on Discrimination

Some have argued that increased reliance on automated reviews of credit has thepotential to reduce discrimination in lending because the automation of decision-makingremoves or reduces the influence of subjective bias. Others have argued that the factorsused to determine a credit score may not completely remove bias from approval andpricing decisions. Furthermore, lenders are still free to offer differential levels of



5

assistance in dealing with errors in credit records, or with other issues related to creditscores, such as providing rescoring services. Such discretionary assistance remains apotential source of bias in the approval process whether a consumer is underwritten withan automated system or with manual underwriting. Federal banking regulators doconduct examinations to ensure against overt discrimination on prohibited bases such as

race, sex, marital status, or age in credit score design or in lenders’ application of thosescoring systems, such as through the use of overrides4.

D. Statistical Validity

Supporters of credit scoring note that credit scores have statistical validity, and arepredictive of repayment behavior for large populations. However, this does not meanthat credit data are error free, nor that credit scoring models are perfect predictors of individual creditworthiness; it only means that they work on average. While the systemsdo present an accurate risk profile of a large numbers of consumers, data users whomanage large numbers of accounts priced by credit risk have a greater tolerance for errors

in credit scoring systems than consumers do. Among those consumers who areinaccurately characterized, businesses can balance errors in their favor against errors infavor of consumers; so long as enough consumers are charged higher rates based oninflated risk assessments to cover the losses from those who are charged lower ratesbecause the systems incorrectly identified them as low risk, these businesses will sufferno material harm. Consumers on the other hand do not have a similar tolerance for errorsin transactions governed by credit reports and credit scores. If they are overchargedbecause of an error in the credit scoring system, there is no countervailing rebate to setthe statistical scales even. Credit scores should not function as a lottery in which someconsumers “win” by being viewed more favorably than they deserve to be, while others“lose” by being viewed less favorably than they should be.

While debate surrounding the broad implications of credit scoring continues, its use isalready strongly established in the American financial services industry. Meanwhile,concern over the integrity of credit scoring itself focuses on two dimensions – the fairnessof the models that interpret the data and the accuracy of the underlying credit related data.

E. Untested Scoring Formulas

Even if all credit data regarding consumers held at credit repositories were accurate,complete, and current, there would be significant concerns about the fairness of automated credit scoring programs. Converting the complex and often conflictinginformation contained in credit reports into a numerical shorthand is a complex process,and requires a significant number of interpretive decisions to be made at the design level.From determining the relative influence of various credit-related behaviors, to the processused to evaluate inconsistent information, there is a great potential for variance amongscoring system designs.

4 See for example Appendix B of the Office of the Comptroller of the Currency’s Comptroller’s Handbook

for Compliance, Fair Lending Examination Procedures, available athttp://www.occ.treas.gov/handbook/fairlep.pdf



6

Despite the gatekeeper role that these scoring systems play regarding access to credit,housing, insurance, utilities, and employment, as well as pricing for those essentials,exactly how the formulas perform the transformation from credit report to credit score isa closely guarded secret. For consumers, regulators, and even industry participants who

rely on the computations in their decision-making, the scoring models largely remain a“black box.” No scholarly reviews of this extremely powerful market force have beenpermitted, and apart from reviews by federal banking regulators to protect againstdiscrimination no government regulator has insisted that they be examined to ensure thatthey are adequate and fair.

Recently, after California passed a law requiring all consumers in the state to have accessto their credit scores, several companies, including Fair, Isaac, and Company, Equifax,Experian, and Trans Union, Fannie Mae, and Freddie Mac have voluntarily providedgeneral information about the information that is used to calculate a credit score or toevaluate a mortgage application, and how that information is generally weighted. In

addition, for a fee, consumers can access score simulators that give some approximationof the impact of various behaviors on their credit scores.

F. Inaccurate credit reports

The most fundamental issue connected to credit scoring is the level of accuracy of theinformation that forms the basis for the scores. Regardless of whether lending andpricing decisions are made by a manual or automated review of a consumer’s credit, thepotential for inaccuracies in credit reports to result in loan denials or higher borrowingcosts is a cause for concern. Several organizations have conducted studies and surveys toquantify the pervasiveness of credit report errors, with widely ranging findings regardinghow many credit reports contain errors (from 0.2% to 70%).

A 1998 study by the Public Interest Research Group5 found that 29% of credit reportscontained errors that could result in the denial of credit (defined as false delinquencies, orreports listing accounts or public records that did not belong to the consumer). The studyalso found that 41% of reports had incorrect demographic identifying information, and20% were missing major credit cards, loans, or mortgages. In total, 70% of reportscontained an error of some kind. This study asked 88 consumers to review their creditreports from each of the three major credit repositories for errors. A total of 133 reportswere reviewed.

Consumers Union has conducted two surveys of credit reports in which consumers wereasked to review their credit reports for accuracy. A 1991 survey6 found that 20% of credit reports contained a major inaccuracy that could affect a consumer’s eligibility forcredit, and 48% contained inaccurate information of some kind. In addition, almost half of survey respondents found that their reports omitted some of their current accounts. In

5 Mistakes Do Happen. Public Interest Research Group. March, 1998.6 “Credit Reports: Getting it Half Right.” Consumer Reports. July, 1991. p. 453.



7

this survey, 57 consumers reviewed total of 161 reports. A 2000 survey7 found that morethan 50% of credit reports contained inaccuracies with the potential to result in a denial,or a higher cost of credit. The errors included mistaken identities, misapplied charges,uncorrected errors, misleading information, and variation between information reportedby the various credit repositories. These results reflect the review of 63 reports by 25

consumers.

A 1992 study conducted by Arthur Andersen8, commissioned by the Associated CreditBureaus (now known as the Consumer Data Industry Association) used a differentmethodology to conclude that the error rate was much lower. This study reviewed thebehavior of 15,703 consumers who were denied credit based on a credit grantor’s scoringsystem. From this sample, 1,223 consumers (7.8%) requested their credit report from theissuing credit repository, and 304 consumers (1.9% of the total sample) disputed theinformation on the report. Of these, 36 disputes (11.8% of those who disputed, or 0.2%of the total sample) resulted in reversals of the original credit denial.

A 1994 study conducted by the National Association of Independent Credit ReportingAgencies (now known as the National Credit Reporting Association) represents a thirdapproach to the question of credit report accuracy. Examining a total of 1,710 files, thisstudy reviewed a three-repository merged infile (which contains the credit reports fromall three credit repositories), and conducted a two-repository Residential Mortgage CreditReport, or RMCR (in which all conflicting data in the two credit repository reports andthe application form is verified with each creditor, and a consumer interview isconducted) for each file. The results showed missing, duplicated, and outdatedinformation in credit files. Among the three-repository merged infiles: 29% of accounts,also known as trade lines or trades (past and current loans, lines of credit, collections,etc.), were duplicates, 15% of inquiries were duplicates, 26% of public records wereduplicates, 19% had outdated trades, and 44% had missing information, such as balanceor payment information. Among the RMCRs: 19% had trades added based oninformation from the loan application, 11% had trades added based on investigations,16.5% had derogatory information deleted as a result of the investigation, 3% had tradesremoved because they did not belong to the borrower, and 2% had errors in publicrecords corrected.

7 “Credit Reports: How do potential lenders see you?” Consumer Reports. July 2000. P. 52-3.8 Described and cited in Klein, Daniel, and Jason Richner. 1992. “In Defense of the Credit Bureau.” Cato

Journal. Vol 12. Issue 2. pp. 393 - 411.



8

IV. How Does the System Work?

The complex system for reporting and reviewing credit involves a large number of participants who fall generally into one of six categories: consumers; data repositories;data users; data furnishers; credit reporting agencies; and analytical service providers.

Approximately 190-200 million consumers have credit reports maintained by the threemajor credit repositories (Experian, Equifax, and Trans Union)9. Data users includelenders, insurers, landlords, utility companies, and employers, who review the creditinformation in consumers’ credit reports to make decisions about extending and pricingcredit, offering and pricing insurance policies, and providing utility services, rentalhousing, or offers of employment. Some, but not all, data users are also data furnishers,and regularly report information about consumers’ accounts to the credit repositories,who add the information to consumers’ credit reports. It is the understanding of theresearchers that there is currently no legal requirement that any business reportinformation to any credit bureau, although once a business furnishes data, there may becertain obligations that arise in connection with consumer disputes. In 1996, Congress

recognized that errors by data furnishers contributed to credit reporting problems, so theFair Credit Reporting Act was amended to impose accuracy duties on data furnishers.These duties are generally subject only to administrative enforcement under the FCRA,with no private right of action for consumers unless the data furnisher fails to complywith re-investigation duties.

Generally, insurers, landlords, utility companies, and employers do not provide positiveaccount information to repositories, nor do all lenders. Also, data enters consumers’records from collection agencies that report on the status of accounts in collection, and

9 Credit repositories attempt to maintain the following information in their databases, but not all data is

available or provided for every account, and different repositories may collect different levels of information, especially consumer identifying information:Consumer identifying information (Consumer’s name; social security number; date of birth; formernames or aliases; current and former addresses; employer; income; position; and employer’s address)Public records information (source of information; date recorded; amount of liability; type of record (e.g.

judgment, tax lien, or bankruptcy); docket number)Collections information (collections company’s name; date opened; last date verified or updated bycollections company; date closed; the amount placed for collection; balance outstanding; name of originalcreditor; the method of payment (a numerical code indicating if the account is current, late, in collection,etc.); any remarks)Creditor information (creditor’s name; account number; level of responsibility for consumer to payaccount (primary account holder, joint account, authorized user, etc.); type of loan (revolving, installment,mortgage, line of credit, etc.) or collateral for an installment loan; date opened; date of last activity; date

closed or paid; highest amount ever owed by consumer; the credit limit on the account; the balance due;payment size and frequency; any amount past due; date of maximum delinquency; dollar amount of maximum delinquency; payment pattern for last 12-24 months (indicating for every month whether theaccount was paid as agreed, or late, and by how many days); the number of months reviewed; number of times account was late by 30, 60, or 90 days; the method of payment (a numerical code indicating if theaccount is current, late, in collection, etc.); any remarks)Credit Inquiries (list of companies who have requested consumer credit information; date the inquiry wasmade)Any consumer statement, such as an explanation of a dispute



9

from repository searches of public records such as bankruptcies, liens, and judgments. Inaddition, governments may report directly to the repositories if consumers fail to paychild support, have unpaid parking tickets, or have been overpaid for unemploymentbenefits. Credit reporting agencies assist some data users by consolidating informationfrom the three credit repositories, and offering services to verify and update information

in credit reports. Credit reporting agencies primarily facilitate and support the decisionmaking process involved with mortgage underwriting. Credit reporting agencies andcredit repositories both provide credit reports to data users, and are considered “consumerreporting agencies” under the Fair Credit Reporting Act. As consumer reportingagencies, these entities share certain obligations, some of which are described below.Analytic service providers also help data users interpret the information in consumers’files, and include companies such as Fair, Isaac, and Company, which produces analyticaltools that generate credit scores, and the Government Sponsored Enterprises (GSEs)Fannie Mae and Freddie Mac, who produce tools that help lenders interpret creditinformation in conjunction with mortgage applications. Some lenders and mortgageinsurance companies have also created tools that help them interpret credit information

for mortgage applications.

A. Non-Mortgage Credit

When a consumer applies for non-mortgage credit, such as a credit card, unsecured lineof credit, or installment loan (e.g. for an automobile, or furniture), the potential creditor(data user) can request a credit report (with or without a credit score) from one, two, orthree of the credit repositories. A repository that receives such a request will send thecredit report to the potential creditor, and record an inquiry on the consumer’s creditreport. The creditor can use the information in the credit report to help decide whether toextend or deny credit to the consumer, and what the interest rate and other fees will be forthis credit. If the creditor accepts the application, they may then act as a data provider,and report information on the consumer’s payment history to one, two, or three of thecredit repositories. Generally account information can be both positive and negative.On-time payments have a positive influence while late payments have a negativeinfluence. However, the amount of positive influence a consumer receives from a timelypayment may vary based on the type of creditor. For example, timely payments to aprime credit card lender may have a greater positive influence on a score than timelypayments to a lender considered less favorable, such as a furniture or consumerelectronics store. If the creditor denies credit, or offers less than favorable terms, basedon the credit report or score, federal laws require them to make certain disclosures to theconsumer, including the name of the consumer reporting agency that supplied the creditreport and how to contact the agency. For non-mortgage applications the consumerreporting agency is usually a credit repository. Once given this information, theconsumer can contact the repository to request a copy of his or her credit report10. If the

10 However, the report the consumer receives may differ from the report that the lender reviewed. If consumers submit more comprehensive personal identifiers in their request for a report from the creditrepository, they may not see the exact report that was used to underwrite their credit application, especiallyif the underwriter made any errors such as misspellings in the consumer’s name or transposing digits in theconsumer’s social security number, or merely submitted an application with less information about the



10

consumer has suffered an adverse action based on the credit report, the copy must beprovided by the repository free of charge. Consumers who have not suffered an adverseaction can also review their credit reports at any time, but are subject to a fee of approximately $9. Six states (Colorado, Georgia, Maryland, Massachusetts, New Jersey,and Vermont) require repositories to provide credit reports to consumers free of charge

once a year upon request. Also, if a consumer is receiving welfare, is unemployed, orsuspects that he or she is a victim of identity theft, the consumer may obtain a creditreport free of charge. For an additional charge, the consumer can have a credit scorecomputed and included with the credit report under any of these circumstances.

B. Employment and Services Other Than Loans

When a consumer applies for employment, or for a service that reviews credit histories,(such as insurance, an apartment rental, utilities, cell phone accounts) these data usersmay also request and receive a credit report and/or scores from one or more repositories,to be used to evaluate the consumer’s application. Job applicants or employees must

provide consent before a report is pulled, but other users derive a permissible purpose toreview credit from the consumer’s act of submitting an application, except in Vermont,where oral consent is required to review a credit report for credit uses.

However, while these entities will review credit, and approve or deny the applicationbased on the credit report and/or score, they generally do not report positive accountinformation back to the credit repositories. They often, however, indirectly reportderogatory information by placing accounts for collection. Accounts that have beenplaced for collection will be reported to one or more of the credit repositories.

C. Other Data Providers

The reverse is true of collection agencies, which provide information to the repositories,but do not use credit data to evaluate consumer creditworthiness, although they may useinformation in credit reports to locate debtors. Repositories also obtain information byrequesting it from public records and government entities and when certain governmententities report directly to the repositories, such as for delinquent child or family supportpayments, unpaid parking tickets, or overpayments of unemployment benefits.Information from collection agencies and public records is primarily derogatoryinformation, such as when an account was sent to collection, or a bankruptcy was filed,but may also include positive information such as the satisfaction of a bankruptcy or therepayment of a collection, and when such repayments occurred. Because governmententities do not report information about bankruptcies, liens, civil suits, or judgments torepositories, the repositories are responsible for maintaining the accuracy of such publicrecord information in credit records, such as whether a bankruptcy has been satisfied or alien has been released. Any type of collection will have a negative impact on a credithistory, regardless of whether the debt was related to an account for which a credit reportwas used to establish credit (e.g. for loans or utilities, as well as for child or family

consumer’s identity. While there is no legal prohibition on lenders providing consumers with the actualcredit report used in their decision-making process, there is likewise no requirement that they provide it.



11

support or parking tickets). Collections, either from a collection agency or other type of account, and public records will continue to have a negative impact after they have beenpaid or otherwise satisfied, although they will have a less negative impact if they aresatisfied, and will have a less negative impact as time passes.

D. Mortgage Credit

The process is more complex for a mortgage transaction. When consumers apply for amortgage, the mortgage lender (who may be a mortgage banker or mortgage broker) hasa number of options that are influenced by what the lender intends to do with the loanafter the closing. The lender can hold onto the loan and collect mortgage payments fromthe consumer until the loan is paid off (known as holding a loan in portfolio), therebyassuming all the risk for borrowers defaulting, or the lender can sell the loan to thesecondary market. If a loan is sold, the originator loses the access to future profits frommortgage payments, but also, so long as the loan meets all the standards set forth by thepurchaser of the loan, retains no risk should the borrower default. The originator retains

the profits from the cost of the mortgage transaction and underwriting, and has areplenished supply of capital to make other loans. The two primary purchasers of loansin the secondary market are the government sponsored enterprises (GSEs) Fannie Maeand Freddie Mac. Lenders may also seek a government guarantee for the loan throughthe Federal Housing Administration (FHA) or Department of Veterans’ Affairs (VA)programs.

1. Portfolio Loans

If a lender is not planning to sell the loan to the secondary market, that lender will usuallyorder a merged credit report, which incorporates information from all three creditrepositories, including the three credit scores. While a lender will generally use reportsfrom all three repositories to underwrite a loan, it may use a single credit report to offer apre-approval. Also, for second mortgages and lines of credit secured by the home,lenders generally underwrite using one credit report. There is no legal or regulatoryrequirement to use a certain number of credit reports to underwrite a mortgage.However, if a lender wishes to sell the loan on the secondary market, or receive an FHAor VA guarantee on the loan it may be required to follow certain protocols.

A lender planning to hold a loan in portfolio will order a merged credit report with scoresfrom a credit reporting agency, passing on information about the consumer such as name,social security number, current and previous addresses. The credit reporting agency willthen pass on the request to a merging company, which will request credit reports from allthree credit repositories and will compile the information from each report returned tothem, according to their merging logic (a set of automated commands designed toidentify shared information and present the three reports in a summarized format). Theindividual credit reports as they read prior to merging and credit scores are also returnedto credit reporting agency. The credit reporting agency will then supply this informationto the lender.



12

Based on the information in this report, and other information such as the applicant’sincome and the loan to value ratio of the mortgage requested, a lender will decidewhether or not to originate the loan, and at what price (interest rate, points, etc.). Anumber of companies, such as mortgage lenders Countrywide and GE Capital andmortgage insurers PMI Mortgage Insurance Company and Mortgage Guarantee Insurance

Corporation, have developed automated underwriting (AU) systems that can provideautomated evaluations of a loan application based on information from the consumer’scredit report and additional information such as income and loan to value ratio.

If the lender is hesitant to originate a loan because of derogatory information in anapplicant’s credit report, and has reason to believe that it may be incorrect, or outdated,the lender can purchase a reinvestigation of the credit information from the creditreporting agency. This entails contacting original creditors, collection agencies, andgovernment records clerks, to verify and update questionable information contained inthe merged credit file. These services can mean corroborating as few as one entry in acredit file, or it can be a comprehensive review in which every entry with conflicting

information is corroborated. An alternative called a Residential Mortgage Credit Report(RMCR) involves reviewing two or three credit repository reports, verifying allconflicting data in the credit repository reports and the application form with eachcreditor, updating any account with a balance over 90 days old, conducting a consumerinterview, and other verification services. Such services provide more currentinformation to a lender for their consideration when underwriting a mortgage, but they donot alter information maintained by any of the credit repositories, nor do they change aborrower’s credit score11. A credit reporting agency may have greater success obtainingclarification of inconsistencies in an applicant’s record than the applicant would haveacting on his or her own, and the credit reporting agency’s reinvestigation is more likelyto be trusted by the lender than the word of a consumer regarding current status of accounts. This service adds cost to the credit underwriting process (roughly $50-100).For consumers who have credit scores far higher than the requirements to qualify, thiswould be an unnecessary service. However, for those who face loan denial, ordramatically higher borrowing costs because of errors in their reports, the savings overthe life of the loan, or in some cases with a single mortgage payment, could more thancompensate for the increased cost of this reinvestigation. After the reinvestigation, thecredit reporting agency will provide the updated and verified information to a lender whocan consider the information while making the final underwriting decision12.

11

When a reinvestigation produces changes in the information contained in a repository’s credit report, thecredit reporting agency is required to pass the information on to the repository within 30 days. However,once this occurs, there is no requirement that the repository update the consumer’s credit file, nor a timeframe within which they must respond. It would be far better for consumers if the credit repositories wereunder an obligation to update the consumer’s file, or at the very least to respond with the results of theirown reinvestigation within 30 days. In the mean time, the disputed information should be part of the creditreport provided to any data users who request the file as the reinvestigation is underway.12 Lenders are not required to accept the results of a reinvestigation, and the automated underwritingsystems of key secondary market actors Fannie Mae and Freddie Mac do not. Instead they require allchanges to be made through a process known as rescoring, described in greater detail below.



13

2. Loans Sold in the Secondary Market

In the current marketplace, few loans are held in portfolio, especially those loansoriginated by brokers. Instead, many are sold into the secondary market to entities thatbundle large numbers of mortgages into securities that are sold to investors – a process

known as securitization. The major actors in this part of the market are the GovernmentSponsored Enterprises Fannie Mae and Freddie Mac, although a number of large nationallenders also purchase and securitize loans. If mortgage originators can sell a loan, thenthey will have renewed capital to make another loan, and will still have profit derivedfrom the costs charged to the consumer for the transaction. Thus selling a loan into thesecondary market is an attractive option.

Government Sponsored Enterprises (GSEs) Fannie Mae and Freddie Mac have bothdeveloped automated underwriting systems which evaluate mortgage applications basedon the information in credit reports, as well as additional information such as income andloan to value ratio, in a very short amount of time. Lenders can submit a loan application

to these automated underwriting systems prior to approving a loan and receive anindication from the GSE that they will purchase the loan. Each GSE has a differentprotocol for submitting loan applications and for obtaining and using credit histories.

Automated underwriting (AU) systems do not approve or deny loans, but can provide anindication of whether a GSE will purchase the loan, and thereby assume the risk of default with respect to the loan. A lender can override an AU decision and underwritethe loan manually, but if they do so, they must agree to buy back the loan if it defaultsand is found to have violated the purchaser’s loan standards. While a loan with an AUapproval that meets all the purchaser’s standards and complies with the warranties of salecarries no risk for a lender or broker, a loan that has been approved by overriding AUstandards does carry significant risk. Many loans are still manually underwritten, but themajority of applications are reviewed with an automated underwriting system, and thisshare is expected to grow in coming years.

Brokers are the dominant originators of loans, but they do not have the financial reservesof banks, thrifts, and other financial institutions. They rely on being able to sell theirloans almost immediately. This is much more difficult without an AU approval. Also,the efficiencies of credit scoring and automated underwriting have made the loanapproval process so fast for loans with good credit that the additional effort required tocorrect errors, or otherwise revisit the details of the loan file, acts as a substantialdeterrent to mortgage lenders working on these loans. In this market, where recordvolumes of loans are being originated, there is a tremendous incentive to deal only withthe loans that will be approved the fastest – the loans that pass the credit score/ automatedunderwriting test13.

13 The economic pressure on originators to underwrite loans that will require the least amount of work existed prior to the introduction of automated underwriting systems. However, the development of automated underwriting has made the process so quick for some loans that the relative additional timerequired to complete a more complicated loan is proportionally greater. Some have noted that decreasing



14

3. Credit Rescoring

If lenders wish to update or correct information in a credit report, the lender cannot usethe reinvestigation process for portfolio loans outlined above and resubmit the loan

through the automated underwriting systems of Fannie Mae and Freddie Mac. Thereinvestigation process outlined above does not change the data on record at therepositories and only reports that contain credit scores and have been generated at therepository level are acceptable for submission to Fannie Mae’s and Freddie Mac’sautomated underwriting systems. Lenders can choose to manually underwrite the loanand submit it with documentation of the errors in the first credit report.

If a lender is unwilling to underwrite the loan manually, and a consumer can afford towait several weeks, the consumer can submit a dispute directly to the credit repository,and the repository has 30 days to respond to the dispute. However, if the borrowerwishes to correct an error in an expedited time frame, lenders who submit loans through

automatic underwriting systems would have to order a service known as rescoring. In thisprocess, the credit reporting agency will obtain the necessary documentation regardingthe disputed account or accounts and contact the rescoring department within the relevantrepository. This department will verify the information provided to them by the creditreporting agency, either through spot checks, or by verification of every update, within afew days. After this process is complete, a new credit report with new credit scores canbe requested, and the loan can be underwritten with the more current information. Inaddition, the information is changed at the repository level, and will be reflected in futurecredit reports for this consumer. This has recently become a very expensive service for alender to purchase. Since the summer, two of the three repositories have increased pricesfor this service by as much as 400%14.

Regardless of how the underwriting takes place, if the loan is originated, the mortgagelender, or the entity holding and servicing the loan if it is sold, may become a dataprovider. The servicer will report information about consumer’s payment behaviorrelated to their mortgage to one, two, or three of the credit repositories, who will add thisinformation to the credit report.

the time required to underwrite the easiest loans potentially frees underwriters to devote more time to more

difficult loans.14 According to reports from a number of credit reporting agencies, Transunion and Equifax have recentlychanged their pricing. Transunion previously charged $5.00 per account entry, or trade line, regardless of whether the account to be updated was a joint or individual account. As of June of this year, Transunioncharges $20 per trade line to update an individual account, and $25 to update a joint account. Equifax hasrecently increased the cost from approximately $5 per rescore to $15 per tradeline for a joint or individualaccount, or $30 for a same day request. Both repositories have clearly stated that these costs are not to bepassed on to the consumer. It is also of note that these two repositories compete with credit reportingagencies in offering rescoring services, and charge between $8-10 per trade line to lenders who contactthem directly.



15

4. Federal Housing Administration (FHA) and Department of Veterans’Affairs (VA) Loans

Lenders who wish to submit loans for an FHA or VA guarantee must also follow certain

protocols regarding the submission of credit reports, but have a number of options tochoose from. For example, the FHA program accepts either a three repository mergedcredit report, a Residential Mortgage Credit Report (RMCR), or applications processedthrough the automated underwriting systems of Fannie Mae and Freddie Mac. TheRMCR option is required to be made available to consumers who dispute informationcontained in their credit reports15. In addition to the options offered to lenders submittingloans for FHA guarantees, the VA program accepts applications processed through theautomated underwriting systems of PMI Mortgage Insurance Company andCountrywide16.

15 See FHA Lender’s Handbook number 4155.1 chapter 2, section 4 “Credit Report Requirements,” andMortgagee Letters 98-14 and 99-26, available at www.hudclips.org.16 See VA Lender’s Handbook, VA Pamphlet 26-7, available at http://www.homeloans.va.gov/26-7.pdf.



16

V. Study Design

A. Phase One

The first phase of the study consisted of a manual review of 1704 credit files, archived by

credit reporting agencies. These files had been requested by mortgage lenders on behalf of consumers actively seeking mortgages. The three credit reporting agencies thatgenerated these files are located in different regions of the county (West, Midwest, andEast) and serve mortgage lenders in a total of 22 states.

Only archived credit files that had been generated by mortgage lender requests for reportsand scores from all three major credit repositories (Experian, Equifax, and Trans Union)were included in the review. Files were included in the study by reviewing consecutivearchived files dating from June 17 to June 20, 200217.

Ensuring the anonymity of all data collected and examined for this study was a

paramount concern for both CFA and NCRA. The data collection procedures weredesigned with particular care to ensure that no personal identifying information fromthese credit files was recorded for this study. No reports were provided in paper orelectronic form, and no names, social security numbers, account numbers, addresses, orother consumer identifying information was recorded. All comments regardinginconsistencies were recorded in generic form. For example, the fact that digits in asocial security number were transposed in one file would have been recorded, but theactual number would not have been. Similarly, if a consumer’s file showed apparentconfusion between credit data recorded under a consumer’s first name and creditrecorded under the consumer’s middle name, this would have been noted, but the nameswould not have been recorded. While the files were being reviewed, the National CreditReporting Association (NCRA) and the Consumer Federation of America (CFA) took precautions to limit the access to identifying information to the credit reporting agencies’representatives, who worked with a representative from the Consumer Federation of America in each office. The credit reporting agency representative retrieved the files,and conveyed only the relevant generic information verbally to the CFA representativefor recording. As a result, the data examined for this study contains only genericinformation about variations in credit data, but does not link that data to any consumer orconsumers.

For each file, the credit scores from each of the three major credit repositories wererecorded. If a repository returned a report, but the report was not scored, or if therepository could not locate a report for the applicant, this information was also recorded.In addition, researchers noted if a file contained multiple reports from any repository, andrecorded the scores for these reports, if the report was scored. Residential MortgageCredit Reports (RMCRs), for which credit reporting agencies verify and update

17 For agencies that serve multiple time zones, additional measures were employed to include records fromconsumers in all regions. For example, every second file from one agency was reviewed rather than everyfile.



17

information in the credit report, were identified as such18. For joint application files, theapplicant’s and coapplicant’s reports were treated as separate reports. Approximately500 files that contained a credit score from each of the three repositories were recorded ateach agency.

A major focus of the study was for those applicants closest to the boundary between thelower priced prime mortgage lending market and the higher priced subprime mortgagelending market, which, in addition to higher costs overall, exposes borrowers to greaterrisks of predatory lending. A large variance between scores on a consumer’s file is alikely indication of drastically incomplete and/or incorrect information in that consumer’scredit reports, and a cause for concern. For those closest to the boundary between primeand subprime, generally considered to be a credit score of 620, the impact of even smallvariances can be severe and translate directly into a greater financial burden.

Thus, more detailed information about each file was recorded: 1) if the file had widelyvarying scores among repositories (defined as a range of 50 points or greater between the

high and low score); 2) if the file was near the threshold between prime and subprimeclassification with a substantial variance between scores (defined as having a middlescore between 575 and 630, and a range between high and low scores greater than 30points); or 3) if the file was directly at the threshold between prime and subprimeclassification (defined as having a high score above 620, and a low score below 620).For files that met these criteria, the four primary factors contributing to the credit score,provided by each repository as part of the credit report, were recorded.

Finally, if the file met criterion 2 (had a middle score between 575 and 630, and a rangebetween high and low scores greater than 30 points), or if the file had a variation inscores of more than 90 points, the specifics of the three credit reports were reviewed in anattempt to identify any obvious inconsistencies between the repositories. When possible,researchers made a determination based on this review of whether any inconsistenciesseemed likely to be artificially lowering or raising the score reported by one or morerepositories.

B. Phase Two

The goal of Phase Two was to test the representational validity of the findings in PhaseOne by comparing key statistics from that sample of credit files with the same statisticsfor a much larger sample of credit files. Specifically, the goal was to compare the rangeamong credit scores, and the frequency of explanations provided to consumers.

This phase of the study reviewed credit scores and the explanations for those scoresprovided by the repositories for a separate sample of 502,623 archived credit files. Thislarger sample was collected electronically and did not involve a manual review of eachfile. As with the first phase, these files had been requested by mortgage lenders on behalf of consumers actively seeking mortgages, and only credit files generated by a request for

18 Conducting and RMCR does not affect the credit scores, and when in depth reviews of the reports wereconducted on RMCRs, the comments referred to the status of the report prior to updates or verification.



18

the reports and scores from all three major credit repositories (Experian, Equifax, andTrans Union) were included.

If a repository returned an unscored report, or if the repository could not locate a reportfor the applicant, this information was recorded. In addition, the presence of multiple

reports from any repository and the scores for these reports, if scored, were recorded. For joint application files, the applicant’s and coapplicant’s reports were treated as separatereports.

For this phase of the study, the zip code for each file was recorded, as was informationabout the type of services requested for each file, and the version of the scoring modelused to calculate each score. By matching zip codes with states, it was possible todetermine the geography represented by these files. Phase Two analyzed files from everystate and territory in the nation, with a wide distribution of files from all regions. (34%from the Northeast, 27% from the Southeast, 30% from the Midwest, 6% from theWest19, 4% with no zip code information to indicate a state, and 0.08% from U.S.

territories.)

Unlike the files in Phase One, which constitute a snapshot of the profile of consumersseeking mortgage credit over just several days, the files reviewed in Phase Two date fromDecember 8, 2000 to September 20, 2002.

C. Phase Three

Phase Three explored the prevalence of specific errors in a representative sample of credit reports, and attempted to quantify how many files contained inconsistent, missing,or duplicated information. Researchers used a 10% sample of all files reviewed at onesite in Phase One and reviewed account data and public records data for errors of omission (information not reported by all repositories) and errors of commission(inconsistent information between repositories, or duplicated information on a singlerepository).

This phase tabulated how many consumer files were missing accounts on at least onerepository report that appeared on other repository reports, treating accounts of differenttype and status separately. The same criteria used to tabulate missing accounts were usedto tabulate the number of files that contained duplicate reports of accounts on a singlerepository report.

19 The researchers were concerned that there were disproportionately fewer files from the western region,particularly a disproportionately low number of files from California. However, subsequent analysisshowed that key statistics and distribution of score ranges for the files from this region, and from Californiaspecifically, were virtually identical to those for the entire sample. Therefore, the researchers are confidentthat this under-representation is not introducing any bias into the findings. (The regions were defined asfollows Northeast: ME, NH, VT, NY, MA, CT, RI, PA, NJ, DE, DC, MD, WV, VA. Southeast: NC, SC,GA, TN, KY, AL, MS, FL, LA, AR, TX, OK. Midwest: OH, IN, IL, MI, WI, MN, ND, SD, IA, MO, NE,KS. West: AZ, NM, MT, WY, CO, UT, NV, CA, ID, OR, WA, AK, HI. Territories: GU, PR, VI.)



19

The seven types of accounts identified were mortgages, other installment loans, revolvingaccounts, other accounts not in collection, medical collections, child support collections,and other collections or charge offs. The researchers differentiated between the status of each non-collection account on the repository or repositories that did report the account.For accounts other than collections and charge offs (mortgages, other installment loans,

revolving accounts, other accounts not in collection), the researchers differentiatedbetween accounts that had no derogatory information, accounts that had late payments,accounts that had conflicting information regarding late payments on two repositories,and accounts that had inconsistent information regarding default. In addition, researchersnoted if a mortgage had gone to foreclosure, and if a revolving account had been reportedlost or stolen.

Files with duplicate or missing public records were tabulated, differentiating by type andstatus as well. Researchers tabulated missing and duplicate bankruptcy filings, liens, judgments, and civil suit filings, differentiating between two categories of status, thosethat had been filed, and those that had been recorded as released, satisfied, dismissed, or

paid.

In addition to determining the number of files with missing and duplicate accounts, theresearchers tabulated the number of files that contained certain inconsistencies betweenthe three repositories regarding account details for accounts reported by all three. Theinconsistencies of interest were: the number of payments recorded as 30 days late; thenumber of payments recorded as 60 days late; the number of payments recorded as 90days late; the balance reported on revolving accounts or accounts in collection; the creditlimit reported on revolving accounts; the past due amount; the method of payment (acode indicating if the account is currently being paid as agreed, is currently late, was late,but is now paid, etc.); the date of last activity on defaulted accounts; and the type of account. Finally, the researchers tabulated the number of files that reported a defaultedaccount, but did not report the date of last activity on that account.



20

VI. Findings

A. Phase One

1. Almost One in Ten Files was Missing a Credit Score from at Least

One Repository.

Of the 1704 unique files reviewed, 1545 files had at least one score reported from eachmajor credit repository. The remaining 159 reports were excluded from the statisticalanalysis because of one or more missing scores. Table 1 details the status of the filesincluded and excluded from the analysis.

2. A Substantial Number of Files Met the Criteria for Further Review.

Of those 1545 files that had valid scores from each repository, 591 files, or 38%, wereflagged for further review, based on the three predefined criteria outlined in the previoussection and below.

Of the 1545 valid files:1. 453 files, or 29%, had a range of 50 points or more between the highest and

lowest scores.2. 175 files, or 11%, had a middle score between 575 and 630 and had a range of 30

points or more between the highest and lowest scores.3. 250 files, or 16%, had high scores above 620 and low scores below 620.

These numbers do not total 591 because many files met multiple criteria. Table 2provides more detail on the number of files that met each of the criteria.

Table 1. Status of Files Reviewed in Phase One.

1390 Files with exactly 3 repositories scored, with no additional scores or unscored reports

114 Files with 3 repositories scored but with additional scores and unscored reports

41 Files with 3 repositories scored but with additional unscored reports

1545 Subtotal: number of files with 3 bureau scores -- included in analysis

58 Files with only 2 repositories scored*

26 Files with only 1 repository scored*

62 Files with no repositories scored*

13 Duplicate files, test files, or other errors that were thrown out

159 Subtotal: number of files excluded from analysis

1704 Total Files Reviewed

* Unscored files include cases where no file was returned (no hit on information input during request) as well

as cases for which a file was returned but not scored.



21

3. Numerous Files Contained Additional Repository Reports andInformation not Relevant to the Consumer’s Credit History.

Each file examined had been generated from a request for a merged file that included onereport and one score from each repository. However, one in ten files (155 out of 1545)contained at least one, but as many as three, additional repository reports. These reportswere not duplicate copies of reports, nor were they residual reports from previousapplications for credit. These additional reports were returned from the samesimultaneous request that produced the other reports in the file. For 114 of the files withadditional reports, at least one, but as many as three of these additional reports alsocontained a credit score. It was unclear to researchers exactly how various systemswould interpret these additional repository reports.

In some cases, an additional repository report was clearly reporting the credit activity of aseparate person (no accounts from the additional report appeared on the three primaryreports, and vice versa). However, it was very common for the additional report tocontain a mixture of credit information, some of which belonged to the applicant andsome of which clearly did not. In some cases, applicants had split files that appeared tobe the result of applying for credit under variations of their name.

Common reasons for returning additional repository reports included:

? Confusion between generations with the same name (Jr., Sr., II, III, etc.).

? Mixed files with similar names, but different social security numbers.

? Mixed files with matching social security numbers, but different names.

? Mixed files that listed accounts recorded under the applicant’s name, but with thesocial security number of the co-applicant.

? Name variations that appeared to contain transposed first and middle names.

? Files that appeared to be tracking credit under an applicant’s nickname.

? Spelling errors in the name.

? Transposing digits in the social security number.

? An account reporting the consumer as deceased.

Table 2. Number of Files that met Criteria for Further Review in Phase One

Met Criterion 1 453

Met Criterion 1 only 273

Met Criteria 1 and 2 only 29


Met all three Criteria 72

Met Criterion 2 175

Met Criterion 2 only 39Met Criteria 1 and 2 only 29



Met Criterion 3 250

Met Criterion 3 only 64




Met any of the three Criteria 591



23

5. Reports Contained Limited Information to Help ConsumersUnderstand the Principal Reasons for their Credit Scores.

If a consumer is subject to an adverse action because of information in a credit report,federal laws (the Fair Credit Reporting Act and the Equal Credit Opportunity Act) requirethe lender to make certain disclosures. Adverse actions include, among other things,denial of credit, or denial of favorable terms on credit. The required disclosures includestatements that an adverse action has occurred and that the decision was based in part orentirely on a credit report and the specific, principal reasons for the adverse action(generally four reasons are given)22.

Thus, each repository report contains the four principal reasons contributing to the scorereturned, as identified by the automated process that calculated the score. The threerepositories have approximately forty standard reasons that can be provided through this

process. However, a mere four reasons were provided as the primary contributing reasonon 82% of the reports reviewed (i.e. the reports in the 591 files that met any of the criteriafor further review outlined in the study design). The four most frequently returnedexplanations for a consumer’s score, with the frequency with which they occurred, were:

22 National Consumer Law Center, Fair Credit Reporting Act, Fourth Edition. 2000.

Chart 1. Middle Score v. Range Between Scores

0

50

100

150

200

250

300

400 450 500 550 600 650 700 750 800 850

Middle Score

R a n g e B e t w e e n H i g h e s t a n d L o w e s t S c o r e s

Regression

trendline

shows slight

correlation

between

middle

score and

range.

A middle score of 620 is

a common dividing linebetween prime and

subprime loans.



24

? “Serious delinquency, and derogatory public record or collection filed” (37% of all explanations).

? “Serious delinquency” (20% of all explanations).

? “Proportion of balances to credit limits is too high on bank revolving or otherrevolving accounts” (15% of all explanations).

? “Derogatory public record or collection filed” (10% of all explanations).

It is important to note that three of the explanations (“Serious delinquency,” “Derogatorypublic record or collection filed,” and “Serious delinquency, and derogatory public recordor collection filed”) convey at least partially redundant information. These threeexplanations alone constituted 67% of all primary reasons provided.

6. In Depth Reviews Revealed Significant Errors and Inconsistencies,Some of Which were Likely Artificially Lowering Consumer CreditScores, and Some of Which were Likely Artificially Raising ConsumerCredit Scores.

In depth reviews were done of files that met the second criterion for further review (had amiddle score between 575 and 630 and a range between high and low score of more than30 points), or if the file had a range between scores of more than 90 points. In each case,researchers attempted to identify any obvious inconsistencies between the account leveldata on each of the repository reports, determine whether these inconsistencies were theresult of omissions, or if they reflected conflicting credit data, and make a determinationof whether the scores were likely being artificially inflated or artificially deflated by theseinconsistencies.

There are obvious limitations to what the researchers could conclude during in depth

reviews of credit file details without the aid of either creditors or consumers tocorroborate or contest inconsistencies. The researchers attempted to approach theseevaluations in as conservative a manner as possible; for example when derogatoryinformation, such as a collection, was reported on only one repository, researchers tendedto assume that the derogatory information was correct. However, when finer details wereinconsistent, such as the current payment status of a given account, the more recentinformation was usually assumed to be correct. In total, 258 files were reviewed indepth.

For approximately half of the files reviewed in depth (146 files, or 57%), researcherswere unable to identify clearly whether inconsistencies in the reports were resulting in an

artificially higher or artificially lower score. In many cases this was because there werelarge numbers of derogatory accounts, reported in various combinations by one, two, orthree of the credit repositories. For those files for which a determination was made, aneven split existed between files for which one or two scores were likely artificially high(56 files, or 22%) and files for which one or two scores were likely artificially low (56files, or 22%). Thus, at least one in five at risk borrowers, but likely many more, arelikely being penalized because of an inaccurate credit report or credit score. Similarly, atleast one in five at risk borrowers is likely benefiting from inflated scores because of



25

incomplete credit information. However, these figures are based on the assumption that,in the absence of contradictory information, all information that was reported by only onerepository was accurate. The figures likely underestimate the actual number of borrowerswho are at risk because they do not account for information that is simply incorrect, doesnot belong to the borrower, or has been contested and removed from one or two

repositories, but not from all three.

While this finding suggests a certain statistical equilibrium between the harm and benefitthat obvious omissions, mistakes, and inconsistencies may be causing to consumers onthe macro level, credit scores are purported to offer consumer-specific evaluations, andare used to generate customer-specific prices and decisions. Lenders suffer little harm solong as there is such statistical equilibrium because the large number of consumers theyserve allows them to benefit from the countervailing impact of these errors on a givenpool of loans. Consumers, on the other hand, have one score for every purchase, and donot benefit from such statistical averaging. Given the number of decisions regardingaccess and pricing of essential services that rely on these scores, their determination

should not be a lottery in which some consumers “win” because derogatory informationis omitted while other consumers “lose” because erroneous, contradictory, outdated, orduplicated information is reported in their credit history. Rather, scores should bedetermined fairly and based on complete, current, and accurate information.

B. Phase Two

The second phase of the study examined the scores and primary factors contributing tothe score, as identified by the repositories, from 502,623 files compiled from electronicrecords. Examining this very large sample allowed for a corroboration of some of thefindings of Phase One among a larger population, roughly equivalent to a 0.25% sample,or one out of every 400 consumers with credit reports. Furthermore, because no detailsof the report were recorded beyond the credit scores and primary reasons for the scores,zip code data could be included without fear of recording excessive personal identifyinginformation. This allowed for verification that the sample had broad geographicalrepresentation.

1. Scores Reported by the Three Repositories for a Given ConsumerVaried Substantially.

The key findings from Phase Two are very similar to the findings from Phase One. Justfewer than one out of four files (105,324 files, or 24%, compared to 21% in Phase One)could be considered extremely consistent, with a range of 20 points or fewer between thehighest and lowest scores. One in three files (129,284 files, or 29%, compared to 31% inPhase One) had a range of 50 points or greater between scores, and one in twenty-fivefiles (17,626 files, or 4%, compared to 5% in Phase One) had a range of 100 points orgreater between scores.

The average (mean) range between high and low score was 41 (compared to 43 in PhaseOne). The median range between high and low score was 35 (compared to 36 in Phase



26

One). Chart 2 is a histogram showing the share of files for which the range betweenhighest and lowest score fell into 10 point bands up to 150, and the number of files forwhich the range exceeded 150.

2. Reports Scored With Different Versions of Scoring Software Reflected

Almost No Difference in Overall Variability of Credit Scores.

As mentioned in the findings for Phase One, some have suggested that score variabilitycan be explained by the fact that different versions of the Fair, Isaac, and Companyscoring software may be in use in the marketplace as data users transition to a newversion. The data collected in Phase Two allowed researchers to assess this anddetermine that the fact that reports were scored with different versions of the scoringmodels did not have an impact on the overall variability of credit scores in this study.

Fair, Isaac, and Company produces the software for all three repositories, but eachrepository refers to the scoring software by a different name. When Experian adopts a

new version of the software, they discontinue the previous version (for example whenthey switched from a version Experian referred to as “Fair Isaac” to a version Experianreferred to as “Experian/Fair Isaac Risk Model”), but users of Trans Union and Equifaxsoftware must update to the newest software version themselves, and there can be morethan one version of the software in use at a given time. The sample examined in PhaseTwo reflected the use of two different versions of scoring software to score reports fromTrans Union and Equifax. Trans Union reports were scored by an older version titled

Chart 2. Frequency of Ranges Between High and Low Score for Phase Two

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

14.00%

16.00%

18.00%

20.00%

1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 101-

110

111-

120

121-

130

131-

140

141-

150

150+

Point Range Between High and Low Scores

S h a r e o f F i l e s W i t h R a n g e



27

“Empirica” and a newer version titled “New Empirica.” Equifax reports were scored byan older version titled “Beacon” and a newer version titled “Beacon 9623.”

The use of different scoring models had a nearly imperceptible effect on variation amongscores. Only three combinations of scoring models occurred in the sample. Reports

scored with the two older versions, “Empirica” and “Beacon,” had an average rangebetween the highest and lowest credit score of 39.61 points, and a median range of 33points. Reports scored with “Empirica” and “Beacon 96” had an average range of 40.85points, and a median range of 34 points. Reports scored with “New Empirica” and“Beacon 96” had an average range of 41.59 points, and a median range of 36 points.Comparing these statistics to the overall statistics for Phase Two (an average range of 41points and median range of 35 points) shows that the influence of different scoringmodels is negligible, and if anything, the newer models resulted in a slightly greatervariation among scores.

Recent commentary suggests that a new version of the software, “ Next Generation

FICO,” which Equifax will refer to as “Pinnacle,” Trans Union will refer to as“Precision” and Experian will refer to as “Experian/ Fair Isaac Advanced Risk Score,”may produce significantly different scores from earlier models, but has not been widelyadopted in the marketplace24. The impact of this new scoring tool is deserving of attention. However, none of the reports in this analysis were scored with this version of the scoring software.

3. Reports Contained Limited Information to Help ConsumersUnderstand the Principal Reasons for their Credit Scores.

As in Phase One, a very limited number of standardized responses represented the vastmajority of all explanations provided to consumers about their credit scores. The samefour explanations that were predominant in Phase One were predominant in Phase Two,but in Phase Two a fifth code was returned with significant frequency.

Three explanations (“Serious delinquency,” “Derogatory public record or collectionfiled,” and “Serious delinquency, and derogatory public record or collection filed”)represented 50% of the primary explanations provided (compared to 67% in Phase One).The explanation “Proportion of balances to credit limits is too high on bank revolving orother revolving accounts” represented 18% of the primary explanations provided(compared to 15% in Phase One). While these explanations constituted a very largeshare of all the principal explanations (7 out of 10), a fifth explanation also constituted asignificant share. The explanation “Length of time accounts have been established”represented 8% of all the primary explanations provided (compared to 5% in Phase One).

23 In addition, 0.3% of files scored by TransUnion were scored by a version titled “Horizon,”approximately 6% of files scored by all three repositories did not identify the version of the software usedfor scoring, and an extremely small number of files (approximately 0.03%) were scored by a non-mortgagemodel, such as an auto model or a bankruptcy model.24 Harney, Ken. “Get Upgraded Credit Scoring,” Washington Post, November 23, 2002, and “Lenders Slowto Adopt New FICO Scoring Model,” Washington Post, November 30, 2002.



28

It is worth noting that the four principal reasons for credit scores were on every fileincluded in the analysis in Phase Two, while Phase One only recorded the explanationsfor those that met the criteria for further review.

C.

Phase Three – Specific Types of Errors

The dramatic ranges between credit scores uncovered in Phases One and Two seem toindicate wide ranging inconsistencies between the information on each repository for agiven consumer. Phase Three attempted to quantify how many consumer files containerrors, and of what kind. Errors of omission (information not being reported by allrepositories) and errors of commission (inconsistent information between repositories, orduplicated information on a single repository) were both considered. Researchersrecorded how many consumer files contained at least one of each category of errorsidentified.

Phase Three re-examined a 10% randomly selected sample of the files reviewed at one of the sites from Phase One. In this sample of 51 three-repository merge files, errors of omission and commission were both rampant. Table 3 lists the categories of errors, thenumber of files that contained such errors, and the percentage of files that contained sucherrors.

This examination of the frequency with which certain errors occur is not intended toimply that the occurrence of any one of these errors alone will necessarily reclassify aconsumer into a more expensive pricing class. The actual impact of any one of theseerrors will depend upon what other information exists in the consumer’s credit report.Any error with the potential to lower a consumer’s credit score will generally have agreater effect on “thinner” files, or files that have less information. Also, if a report hasno derogatory entries, the first piece of derogatory information will very likely have amore severe negative impact on a consumer’s apparent creditworthiness than the sameinformation would have on a file with multiple derogatory entries. However, it ispossible for a single derogatory entry to have a dramatic effect on a consumer’s score,whether or not it is accurate. If that consumer is near the threshold for a less favorablepricing class, it is very possible and probable that an error or errors in that consumer’scredit history could have a substantial material impact. Furthermore, most reportsreviewed contained more than a single error, and the cumulative effect of multiple errorsincreases the likelihood of material impact on consumers.

The sample size in Phase Three is the smallest of the three phases, due primarily to thetime required to review files in sufficient depth to identify specific errors. Theresearchers recognize that the statistics from this phase have limitations and it is difficultto make definitive statements about the frequencies with which specific errors occur inthe population at large based on these findings. However, this phase does documentstrikingly high levels of errors and provides evidence that at the very least a significantminority in the general population are at risk for a variety of errors of commission andomission.



29

Table 3. Types of Errors, and Number and Percentage of Files Containing Such Errors

Type of Account Status N u m b e r o f F i l e s M i s s i n g

S u c h A c c t .

% o

f F i l e s M i s s

i n g S u c h

A c c t .

N u m b e r o f F i l e s w i t h

S u c h A c c t . D u p

l i c a t e d

% o

f F i l e s w i t h

S u c h

A c c t . D u p l i c a t e

d

N u m b e r o f F i l e s w i t h

I n c o n s i s t e n t I n f o

% o

f f i l e s w i t h

I n c o n s i s t e n t I n f o

Mortgage No Derogatory Info 17 33.3% 1 2.0%

Mortgage Late Payments 1 2.0% 0.0%

Mortgage Inconsistent Lates btw Repositories 1 2.0% 0.0%

Mortgage Inconsistent, one shows Default 0.0% 0.0%Mortgage Foreclosure 2 3.9% 1 2.0%

Other Installment No Derogatory Info 34 66.7% 4 7.8%

Other Installment Late Payments 3 5.9% 0.0%

Other Installment Inconsistent Lates btw Repositories 2 3.9% 1 2.0%

Other Installment Inconsistent, one shows Default 1 2.0% 0.0%Revolving No Derogatory Info 40 78.4% 9 17.6%

Revolving Late Payments 6 11.8% 0.0%

Revolving Inconsistent Lates 2 3.9% 0.0%

Revolving Inconsistent, one shows Default 4 7.8% 0.0%Revolving Missing Lost or Stolen 8 15.7% 0.0%

Other No Derogatory Info 8 15.7% 1 2.0%

Other Late Payments 0.0% 0.0%

Other Inconsistent Lates btw Repositories 0.0% 0.0%Other Inconsistent, one shows Default 0.0% 0.0%

Collection Medical Collection/ Chargeoff 10 19.6% 0.0%

Collection Child

Support Collection/ Chargeoff 1 2.0% 0.0%

Other Collection orChargeoff Collection/ Chargeoff 13 25.5% 3 5.9%

Bankruptcy Filed 0.0% 0.0%Bankruptcy Released/Satisfied/Dismissed/Paid 5 9.8% 1 2.0%

Lien Filed 4 7.8% 0.0%Lien Released/Satisfied/Dismissed/Paid 2 3.9% 0.0%

Judgement Filed 3 5.9% 0.0%Judgement Released/Satisfied/Dismissed/Paid 2 3.9% 0.0%

Civil Suit Filed 0.0% 0.0%Civil Suit Dismissed 1 2.0% 0.0%

# 30 Late 22 43.1%

# 60 Late 15 29.4%# 90 Late 12 23.5%

Balance on Revolving Accts or

Collections 42 82.4%

Credit Limit on Revolving Accts 49 96.1%

Past Due Amount 9 17.6%Current Method of Payment 31 60.8%

Type of Account 11 21.6%Last Activity on Defaulted 13 25.5%

No Last Activity Date on defaultedaccounts 11 21.6%

Omission Commission



30

1. Significance and Frequency of Errors of Omission

Incomplete reporting of information, or an error of omission, can make a consumer

appear either more credit worthy or less credit worthy, depending on the nature of theinformation that is omitted. When a derogatory account, such as a collection, latepayment, charge off, or public record is omitted, the consumer’s record will appear lessrisky, and the consumer’s credit score will likely be artificially high. However, when apositive account, such as a mortgage, auto loan, or credit card account that has been paidas agreed, is omitted, this responsible credit behavior will not be conveyed and theconsumer’s credit score will likely be artificially low.

Positive account information is especially important for consumers who are justbeginning to establish credit, or who are working to re-establish their credit rating afterbankruptcy. Omitting positive information can have a dramatically negative impact on

such consumers. Failure to report positive accounts can deflate scores, or even make itimpossible for the scoring model to produce a score. Such outcomes make it moredifficult to enter or return to the prime lending marketplace, relegating affectedconsumers to the higher priced subprime market.

Because of the limitations of the study, researchers were unable to determine definitivelywhether many of these errors were errors of omission. For example, researchers couldnot be certain that accounts appearing on one report only were the result of omissions bythe other two repositories, or if the accounts appeared as the result of merging errors, orcompiling errors on that one repository (and actually did not belong to the consumer), orif they had been contested and removed from some repositories but not removed from allthree. In the absence of evidence that presented a contradiction, researchersconservatively treated information appearing only on one or two repositories as an errorof omission.

a) More Files Contained Omissions of Positive Information thanContained Omissions of Derogatory Information, but Omissions of All Kinds were Common.

Accounts that had never been late, and which have great significance for determining acredit score, were omitted with extremely high frequency. Omitted revolving accountswith no derogatory information were noted on the largest number of consumer files.Nearly eight out of ten files (78.4%) were missing a revolving account in good standing.In addition, one file out of three (33.3%) was missing a mortgage account that had neverbeen late, and two files out of three (66.7%) were missing another type of installmentaccount that had never been paid late. Other accounts with no derogatory information,such as non-revolving credit cards, were missing on 15.7% of all files.

Omissions of accounts with late payments, but which had not been sent to collection,were less frequent than omissions of positive accounts. Still, one in ten files (11.8%),



31

was missing a revolving account with late payments reported, and many (7.8%) weremissing revolving accounts that were being reported as defaulted by one of the tworepositories that reported the account. Half that number (3.9%) contained conflictinginformation about late payments on revolving accounts reported by two repositories. Amuch smaller number of files were missing mortgages or installment accounts that had

been late at some time in the past, or that had conflicting information regarding latepayments, but 3.9% of files omitted a foreclosure.

The most commonly omitted derogatory information was for various types of collections.Child support collection omissions were rare (2% of files), but one out of five files(19.6%) omitted a medical collection, and one out of four files (25.5%) omitted acollection of some other kind.

b) Medical Collections Raise Special Concerns RegardingAppropriateness and Privacy.

Medical collections, as a subset of collections that were often not reported on all threerepositories, deserve special attention. Disputes between consumers, health insurancecompanies, and medical care providers occur frequently, and can be of extended duration.Many medical bills are referred to collection agencies during these disputes but areultimately paid by insurers. Therefore, if all the relevant facts were known thesecollections could very likely be errors of commission, rather than errors of omission, asthey may not accurately reflect consumer debt repayment behavior.

Another issue noted by researchers related to medical collections was the high degree of information that can be inferred from the information in medical collection entries listedon a consumer’s credit report. The names of many medical creditors are specific enoughto allow for identification of categories of treatment. For example, information incollection entries identified categories of medicine, such as perinatology, and neonatalhealth clinics. This could have especially significant ramifications if full credit reportsare reviewed by potential and current employers, who may infer from such collectionsthat an applicant, or employee, has an unusually sick newborn, and may be more likely tobe called away from the office25. In other cases, consumers may simply wish not to havethe fact that they have sought treatment for other very private matters (such as treatmentsfor fertility, mental health, or AIDS) to be readily discernible by anyone who reviewstheir credit record.

Section 604 (g) of the Fair Credit Reporting Act states that “ A consumer reportingagency shall not furnish for employment purposes, or in connection with a credit orinsurance transaction, a consumer report that contains medical information about aconsumer, unless the consumer consents to the furnishing of the report.” However,consumers have complained about the difficulty of identifying the original creditors forcollection accounts that appear on their files, and best practices have been proposed by

25 It is the researchers’ understanding that current market practices do not permit employers to view thesame level of detail that is provided to potential lenders. Employer credit reports generally do not containthe notations on collection entries that would allow them to make such medical inferences.



32

the Consumer Data Industry Association that attempt to strike a balance betweenprotecting consumers’ medical information and providing enough information to allowconsumers to identify the original source of debts. Furthermore, it is the Researchers’understanding that in Massachusetts, the original creditor must be listed for everycollection account.

c) Public Record Information was Frequently Omitted, IncludingBoth Information that Would Likely Increase Credit Scores andInformation that Would Likely Decrease Scores.

One in ten files had an omitted date of fulfillment for a bankruptcy, an omission thatalmost certainly lowered the corresponding credit scores. Several files also containedreports that omitted liens, both satisfied (3.9%) and unsatisfied (7.8%), and judgments,both satisfied (3.9%) and unsatisfied (5.9%). One file contained a dismissed civil lawsuit that was reported to one repository only.

Given the dramatic frequency of omissions of both positive information (such asmortgages) and derogatory information (such as collections and public records) it is clearthat errors of omission have the potential to undermine the accuracy of consumer creditrecords and, by extension, credit scores. It should be noted that true errors of omission(excluding unrelated account information that is erroneously captured by one repositoryand disputes which have not resulted in removal of information from all threerepositories) are most likely the fault of the creditor, not the credit repository. If a dataprovider, be it a collection agency or major national bank credit card, decides not toreport information to all three repositories, then the repositories do not know theinformation and cannot report it.

2. Errors of Commission

Also of great concern to consumers is the frequency with which errors of commission, orinclusion of incorrect information, occur in credit reports. A credit report with incorrectderogatory information makes a consumer appear to be a greater lending risk and willlikely artificially lower the consumer’s credit score. In addition, duplicate reporting of accounts can have an impact on a consumer’s scores.

Again, because the researchers did not have the benefit of knowing the consumers’ credithistories, we were limited in the errors of commission that we could identify. Only incases where repositories were reporting conflicting details on an account couldresearchers identify with certainty that at least one repository was incorrect. Even withthese limitations, the findings are troubling.

a) Many Consumer Files Contained Conflicting InformationRegarding the Consumer’s Record of Late Payments.

In 43.1% of the files, reports regarding the same accounts conflicted regarding how oftenthe consumer had been late by 30 days. In nearly one out of three cases (29.4%), there



33

was conflicting information about how many times the consumer had been 60 days late,and conflicting information regarding the number of times an account had gone to 90days late in one out of four consumer files (23.5%). Late payments, especially on recentaccounts, can be very detrimental to a consumer’s credit score. Delinquencies areidentified as major contributing reasons for a consumer’s score on the majority of reports.

In some cases, but by no means in all, different numbers of late payments may be theresult of the timing of record updating procedures by the repositories. For example, onerepository may have information on an account that is current as of June, whereas anotherrepository may only have received or loaded information current as of May. However,this phenomenon would only explain variations for accounts that are currently past due,and not for the significant number of files that were currently reported as paid on time,but had discrepancies in the historical count of late payments. Furthermore, regardless of a repository’s particular timing, a consumer will be evaluated on the informationavailable at the time of application.

b) Reporting of Account Balances was Inconsistent

Inconsistencies regarding the balance on revolving accounts or collections appeared on82.4% of files, and inconsistencies regarding an account’s credit limit appeared on 96.1%of files. These particular numbers are presented with one qualification. The softwareused to review reports presents information in a field titled “credit limit/high credit.”Researchers acknowledge that the raw data may contain separate information regardingthe high credit (the highest amount ever charged on this account) and the credit limit (theamount of credit made available by the creditor) and the observations regardinginaccuracies in these fields may not reflect the data used to derive credit scores.However, even with this qualification, there are reasons to be concerned about incorrectreporting of balances or credit limits. Credit card lenders have an incentive to obscurethe real credit limit from credit reports, as a means of retaining existing borrowers. If acredit card lender reports a credit limit as lower than the actual limit (for example byreporting the high credit as the credit limit) the borrower will appear to be closer to“maxing-out” their credit, and will appear less attractive to competing credit card lenders.Thus, the consumer will be less likely to receive competing offers. Such misreportingalso poses a significant risk to consumers’ overall credit rating. The practice of deliberately refusing to report complete and accurate account information in orderobscure consumers’ credit has drawn repeated condemnation from John Hawke, theComptroller of the Currency26. There is good reason to be concerned, given that one of

26 In a May 5, 1999 speech before Neighborhood Housing Services of New York, Hawke stated, “Subprimeloans can’t become a vehicle for upward mobility if creditors in the broader credit market lack access toconsumer credit history. Yet, a growing number of subprime lenders have adopted a policy of refusing toreport credit line and loan payment information to the credit bureaus – without letting borrowers knowabout it. Some make no bones about their motives: good customers that pay subprime rates are toovaluable to lose to their competitors. So they try to keep the identity and history of these customers aclosely guarded secret” (http://www.occ.treas.gov/ftp/release/99-41a.doc). He reiterated these concerns ina June 9, 1999 speech before the Consumer Bankers Association, condemning the objectionable practice of non-reporting and noting that, “failure to report may not be explicitly illegal. But it can readily be



34

the most frequently provided explanations for a consumer’s credit score is that the“proportion of balances to credit limits is too high on bank revolving or other revolvingaccounts.” This is the primary explanation listed on approximately one out of six reports.

c) Contradictory or Missing Dates Occurred Frequently and Have

the Potential to Distort a Consumer’s Record.

Because more recent credit activity is more influential in determining a credit score, it isimportant that the relevant dates on accounts be accurate. This is primarily true foraccounts that have gone into default. Creditors track the date of last activity on consumeraccounts, but, because most creditors report to repositories in large batches of data onmany accounts, credit repositories also track a second date – the last date the informationwas reported by the data provider. If a data provider fails to report any information in thedate of last activity data field, the scoring software will assume that the date last reportedis the date of last activity. Thus, if a consumer has an account that defaulted severalyears ago, but otherwise has good credit, under normal circumstances the relative impact

of this account will diminish over time. However, if there is no date of last activityreported, this default will seem perpetually as recent as the last submission of a batch of data from that provider. One in five consumer files (21.6%) contained a defaultedaccount that did not report a date of last activity. One in four files (25.5%) containedcontradictory information regarding the date of last activity.

d) Duplicate Reporting of Accounts did not Appear to be asWidespread as Many of the Other Errors Noted in thisInvestigation.

When accounts were reported multiple times by a single credit repository, they tended tobe accounts that had no derogatory information, which may provide an artificial boost toa consumer’s credit scores by giving the impression that the consumer has successfullymanaged more credit than he or she actually has, but may also lower a consumer’s creditscore by increasing their apparent overall debt load. Also, on 5.9% of files a collectionwas reported more than once on a single credit report, likely artificially lowering thescore. This was usually the result of a collection being reported by the original creditoras well as a collection agency that had taken over the account.

Further contradictions existed regarding the method of payment (whether an account wascurrent, late, charged off, in collection, etc.) on 60.8% of files, the type of account(revolving, installment, mortgage) on 21.6% of files, and the past due amount on 17.6%of files.

3. Merging and Compilation Errors

Credit data are complex, and accurate interpretation of it can sometimes take aconsiderable amount of time and effort. When credit reporting agencies and credit users

characterized as unfair; it may well be deceptive, and – in any context – it’s abusive”(http://www.occ.treas.gov/ftp/release/99-51a.doc).



35

review merged reports, they employ software to help organize and simplify theinformation, so the user can quickly assess the unique information contained in eachrepository without having to sift through the same information reported by anotherrepository. The design of a tool to do such work involves making certain choices, whichcan lead to significantly different results. For example, some merging software is

designed to present the details for a given account from one of the three repositories to acredit user, and “hide” the other two repositories reports. Other software utilizes amerging logic that takes some information from each repository report to create anamalgam of the information in each credit report. This one example of a design decisioncan result in a very different presentation of the same raw data to a credit reportingagency or credit user.

The discussion of duplicate and mixed files in Phase One already illustrated that a largenumber of errors enter the credit reporting system when the automated software used bythe credit repositories compiles information about credit users. Use of nicknames,misspellings, transposed social security numbers, and mixed files that report information

under one person’s name, but match that name to a spouse’s social security number, areall examples of variations that can result from an automated interpretation of complexand sometimes contradictory personal identifying data. Software designers must makeexplicit choices about how to interpret this data, and what form the output will take. Forone in ten files, the result was an additional repository report and/or an additional creditscore.

A similar potential for error exists when automated systems interpret multiple reports,merging the three credit reports into a single representative report. This process attemptsto reconcile the voluminous inconsistencies between repositories for account levelinformation. Given the difficulties that are apparent from the attempts to reconcileindividual consumer information, the importance of ensuring a fair and rigorous merginglogic for any compilation software is clear.

These concerns raise many questions. How exactly does a software program that collectsinformation from multiple credit repositories interpret conflicting or duplicatedinformation? How much variation can a given software package consider before anaccount entry is treated as a separate account? How many creditors are trying to gamethe marketplace by not reporting complete or accurate information about consumers – ineffect making consumers appear less creditworthy than they actually are to other potentialcreditors, in a bid to protect their customer base?

We do not raise these problems to advocate an end to use of multiple repository reports.In fact, use of multiple credit scores serves as a control against errors of omission. (Allof the errors of omission identified in this study were identified because of the use of multiple repository reports.) On the contrary, we identify these problems to illustrate thatthere are difficult choices that must be made when developing all of the components of the interconnected system that evaluates credit. Given the lack of oversight of thisdimension of the market, there is a very real potential for developers to make choices that



36

result in a system that is unfair to consumers in general or to a certain segment of consumers, such as those nearest the threshold between prime and subprime.



37

VII. Conclusions and Implications of the Findings for Consumers

A. Credit scores and the information in credit reports vary significantly among

repositories.

The scores based on data from the three repositories can vary dramatically for allconsumers regardless of whether they have generally good or bad credit histories.Approximately one out of every three files (31%) had a range of 50 points or greater, andone out of twenty reports had a range of 100 points or greater (5%). The average rangebetween high and low scores was 43 points (median range was 36).

The wide range in credit scores reflects a similarly broad variation in the data containedin each repository report for a given consumer. Significant accounts, such as mortgages,credit cards, collections, and public records, were regularly omitted from one or morecredit repository reports. In addition, for most consumers, the details of accounts that arereported by all three repositories are unlikely to be completely consistent. Information

about late payments, the balance and credit limit on revolving accounts, and the currentstatus of accounts are among errors that occur frequently.

B. Many consumers are unharmed by these variations, and some probably

benefit from them.

Consumers with very good credit histories, whose credit scores place them firmly abovethe cutoff for the most the favorable product terms, are as likely as any other consumer tohave variation between credit scores. However, as long as that variation does not resultin scores that are lower than the qualifying score for the best terms for credit, insurance,or any other product or service underwritten by their credit score, there will be nomaterial harm. The number of consumers in this category is somewhat unclear anddepends upon the products being sought and the qualifying scores for those products.

Furthermore, those near the boundary between pricing ranges, such as the divisionbetween the prime and subprime mortgage markets, who have errors that artificially raisetheir scores may be artificially classified as lower risk. As a result, such consumers havethe potential to reap some benefit from the inconsistencies.

C. However, tens of millions of consumers are at risk of being penalized for

incorrect information in their credit report, in the form of increased costs or

decreased access to credit and vital services.

We estimate that tens of millions of consumers are at risk of being penalized byinaccurate credit report information and incorrect credit scores. Between 190 and 200million Americans, or nearly every adult consumer, has a credit report that can be scoredto produce a credit score. Businesses from mortgage lenders to utility providersincreasingly have established pricing structures in which the charge for the loan orservice corresponds to a credit score range. Errors in credit reports that lower aconsumer’s credit score can place that consumer into a more expensive pricing range than



38

he or she deserves to be in. Credit scores below a certain cutoff point can even disqualifyconsumers outright.

Looking at the mortgage market as an example, the two most significant ranges aredefined by a credit score of 620. Whether a consumer’s credit score is above 620 or

below 620 determines if the consumer qualifies for

27

the lower priced prime market, or if the consumer will be limited to subprime market, which imposes higher borrowing costs,often requires larger down payments, and exposes consumers to abusive predatorylending practices. In addition to this primary division in the prime and subprimemortgage markets, there are secondary pricing ranges. According to the consumerfocused website of Fair, Isaac, and Company (www.myfico.com), consumers with a scorebetween 720 and 850 will qualify for the lowest interest rates, but there are at least fourdifferent pricing ranges in the prime market and at least two in the subprime market.Consumers with a score between 700 and 719 will be charged higher borrowing coststhan those in the highest score range. Prices similarly increase for scores between 675and 699, and between 620 and 674. Within the subprime market, the two pricing ranges

identified by Fair, Isaac, and Company are from 560 to 619 and from 500 to 559.

This study focused on consumers at risk for misclassification into the subprime marketdue to inaccurate information in their credit report and found that one in five consumers(20.5%) is at risk. We have defined at risk consumers as either having a middle creditscore between 575 and 630 with a score variance of greater than 30 points, or as having ahigh score above 620 and a low score below 620. Among these at risk consumers, basedon our analysis of files, we estimate that at least one in five (22%) is likely beingpenalized with lower scores than deserved because of errors or inconsistencies in his orher credit report that are clear enough to be noticed by an outside observer unfamiliarwith that consumer’s debt payment history. (We also estimate that at least one in five(22%) has scores that are likely too high due to a lack of reporting by creditors to allrepositories.) The remaining sixty percent of at risk consumers have credit reportswithout errors clear enough to allow an outside observer to determine whether their creditscores are artificially low or artificially high. We strongly suspect that a significant shareof these at risk consumers also have artificially low credit scores due to errors in theirreports that they would be able to identify if given the opportunity.

While the findings suggest that there may be some statistical equilibrium between thoseconsumers who have artificially high scores and those who have artificially low scores,such statistical averaging is irrelevant to the individual consumer who is penalized basedon errors in his or her credit report. Credit scores are purported to offer consumerspecific evaluations of credit and do result in consumers specific decisions regardingpricing and availability for the essentials of daily life and economic activity.

Consumers may be harmed by both errors of commission and errors of omission. Errorsof commission can lower a consumer’s score in situations such as when incorrect

27 Because of the aggressive sales tactics of subprime and predatory lenders, many consumers who havecredit scores above 620 have subprime loans, although they could have qualified for less expensive primeloans. This is an important but separate issue.



39

information or mixed files add the credit history of others to a consumer’s report. Errorsof omission can lower a consumer’s score when the record does not contain full andaccurate information regarding existing accounts paid as agreed.

Those consumers on the threshold of subprime status face particularly dire consequences

from this lack of precision. Falling below the cutoff score for a prime rate mortgage canadd a tremendous financial burden to these threshold consumers and make it moredifficult to meet this and other financial obligations. Interest rates on loans with an “A-”designation, the designation for subprime loans just below prime cutoff, can be more than3.25% higher than prime loans. Thus, over the life of a 30 year, $150,000 mortgage28, aborrower who is incorrectly placed into a 9.84% “A-” loan would pay $317,516.53 ininterest, compared to $193,450.30 in interest payments if that borrower obtained a 6.56%prime loan – a difference of $124,066.23 in interest payments29.

We conservatively estimate that 40 million consumers (twenty percent of the 200 millionwith credit reports) are at risk of being misclassified into the subprime mortgage market,

and at least 8 million (twenty percent of these at risk consumers) would be misclassifiedas subprime upon application, but the actual numbers are likely much higher. Thesenumbers do not even attempt to quantify the number of consumers who are beingovercharged because errors pushed them into a higher pricing range within the prime orsubprime markets. Furthermore, consumers with errors in their credit reports andartificially low credit scores are penalized in a number of markets in addition to themortgage market. These figures do not address the consumers penalized with highercredit card interest rates, more expensive insurance, or those denied insurance, housing,utility service, or employment (an application of credit scoring we expect to increase infrequency) on the basis of erroneous credit scores.

D. Almost one in ten consumers runs the risk of being excluded from the credit

marketplace altogether because of incomplete records, duplicate reports, and

mixed files.

If a consumer has very little credit history, or is rebuilding credit after a bankruptcy,every positive account that they can establish is vital for creating a record that hassufficient information to be scored. If a lender requests scores for a consumer, but arepository is unable to return a score (as was the case for approximately one out of tenfiles reviewed in this study), that lender may choose to set aside the customer’sapplication and focus on an application with enough credit to be scored and priced withminimal work. This is especially likely during periods of heavy volume, such as theprolonged refinancing boom currently occurring. Even if a lender later returns to the filethat was set aside once volumes have subsided (perhaps because of seasonal fluctuationsin home buying activity, or because interest rates have risen), the consumer will havesuffered substantial harm by being excluded even temporarily from the marketplace.

28 The Federal Housing Finance Board’s Monthly Interest Rate Survey reports that the national average loanamount for conventional home purchase loans closed during June of 2001 was $151,000.29 Interest rates as reported by Inside B & C Lending for 30 year Fixed Rate Mortgages for “A-” Credit (parpricing), and “A” Credit respectively, as of July 14.



40

Consumers may not understand the implications of incomplete reporting or non-reportingby their creditors, and would have little leverage to force their creditors to report up todate information anyway.

Similarly, consumers generally have no control over the inclusion in their credit files of

duplicate reports, or mixed information not belonging to them. The only person in aposition to tell if a credit repository’s compilation system incorrectly groups unconnectedinformation with a consumer, or to assess why their credit record was not scored, is thelender. But there is no requirement that the lender share the report or score with theconsumer. Furthermore, if the lender incorrectly enters the identifying information,during a credit review, either leaving out information such as social security number,generation (Jr., Sr., etc.), or mistyping the applicant’s name or other information, thelender may be contributing to the problem. If a consumer later requests a copy of his orher credit file after denial, he or she will often be required to provide morecomprehensive information than the original data user. This means that the reporteventually provided to the consumer may have a lower propensity of errors than the

version used to evaluate his or her application. This is especially true for non-mortgagecredit, or mortgage credit underwritten with files ordered directly from one or more creditrepositories. If a mortgage lender ordered a merged credit report from a credit reportingagency that merged the files into a new report, and after being denied the borrowerrequests a copy of the credit report from that agency, the agency has an obligation to givethe consumer the merged credit report.

The treatment of unscored files is a very serious question. How do automated creditreviews treat files that contain extra scores, or extra reports that are unscored? One in tenrequests fails to return a score from each repository. As many requests return one scorefrom each repository, but also return additional files that may or may not be scored. If automated credit reviews reject additional files, as many as two in ten consumers couldbe excluded from the credit market outright because of these problems.

E. The use of information from all three repositories in mortgage lending

protects consumers and creditors from being negatively affected by errors of

omission, but it may increase the negative impact on consumers of errors of

commission.

The use of information from all three repositories on mortgage underwriting offersconsumers and creditors protection against errors of omission by introducing themaximum available information to the scoring and underwriting process. However,errors of commission actually occur on more files than do errors of omission, and thereare a number of different approaches to using information from three repositories forunderwriting purposes. Without a chance for borrowers to review their reports for errorsof commission at the time of underwriting, and without oversight of how the informationis merged and presented, the use of multiple repository sources of data can produce aresult that is harmful to consumers.



41

F. Consumers are not given useful and timely information about their credit.

1. Standardized, generic explanations do not provide sufficientinformation for consumers to address inconsistencies and contradictions,let alone outright errors.

Approximately 7 in 10 credit reports indicated that the primary factor contributing to thescore was “serious delinquency, derogatory public record, or collection filed,” or somesubset or combination of these factors, without providing any information about whichspecific accounts were responsible for the low scores. In many cases, it is not even clearwhether a delinquency, public record, or collection was responsible for the score. Inaddition approximately one in six reports indicated that the primary reason for the scorewas that the proportion of revolving balances to revolving credit limits was too high.These two relatively generic explanations were reported as the primary reason for aderogatory score on more than 8 out of 10 reports reviewed.

The vague information provided by these explanations is too general to be helpful. Nearlyall consumers near the subprime border have had some activity in their past that may fallunder the broad terminology “serious delinquency, derogatory public record, or collectionfiled,” almost by definition. If their credit records were more favorable, they would notbe so close to the subprime threshold. Such borrowers may accept this generic justification for a low score more readily than consumers with generally good credit.Thus, the consumers who are most likely to be penalized by errors are the least likely tochallenge these imprecise explanations. Because threshold consumers are not providedthe specific account information that is lowering their scores, they are not given the toolsto identify and correct possible errors. The situation would likely be different if consumers had access to the full credit reports and scores used to underwrite their loanapplications, with an indication of which accounts had the largest negative effect on theirscores. If this were the case, consumers would have a much more legitimate opportunityto identify and challenge any errors.

The credit report is a rare type of consumer product. Consumers pay for it duringmortgage underwriting, and are rewarded or penalized on the basis of it, but are not evenallowed to look at it, much less keep a copy for their records. Furthermore, consumerscan understandably view the report as “theirs” because it is purportedly a record of theirbehavior.

2. Consumers outside of California have no affirmative right to knowtheir credit scores.

Credit scoring is a shorthand that allows lenders to more quickly assess the complexinformation in a consumer credit report. However, with the exception of Californiaresidents, consumers are not guaranteed access to their credit scores, although they arepermitted to purchase copies of the underlying data. Thus, consumers are placed at adisadvantage relative to lenders when it comes to evaluating their own credit-worthiness.When Californians gained access to their scores, many lenders across the country did



42

begin making the scores available. As with the specific credit report used to evaluate anapplication, consumers are charged for the additional cost of obtaining a credit score forunderwriting, but have no guarantee that they will be able to view the specific score usedto underwrite their loan. Currently, all three repositories allow consumers to purchasescores in conjunction with credit reports, but prior to the passage of the California law

requiring this, the repositories resisted providing scores to consumers.

G. Private companies without significant oversight are setting, or at the very

least heavily influencing, the rules of the marketplace for essential consumer

services that base decisions on credit scores.

Companies, such as Fair, Isaac, and Company, have produced credit scoring software thatis increasingly used in the marketplace to determine access and pricing for the essentialsof daily life and economic activity. Consumers have no choice regarding how lenders orother data users evaluate their credit, and widespread and increasing use of credit scoringsystems that evaluate applications for credit, mortgages, insurance, tenancy and even

employment is a fact of the marketplace. Scoring systems incorporate many complexdecisions regarding the interpretation and treatment of information that can becontradictory, incomplete, duplicative, or erroneous. There is great potential for thesesystems to incorporate inappropriate decisions that result in consumer harm, especially asmodels originally designed to evaluate credit applications are adapted to evaluateapplications for services completely unrelated to credit behavior.

Despite the tremendous and growing influence of automated credit evaluations, nogovernment entity has recognized and acted on the clear need for ongoing, timely reviewof these software systems to determine their accuracy, fairness and appropriateapplication. Many decision-makers who use scoring systems to evaluate consumerapplications do not even understand the systems themselves and cannot explain them toconsumers. Thus, while decision-makers are increasingly relying on programs that theydo not understand, no public entity is guaranteeing the validity and fairness of suchprograms. Without independent review and oversight of this market force, consumersare, literally, left to the devices of the system developers.

H. Certain information in credit reports has the potential to cause breaches of

consumers’ medical privacy.

Many credit report entries regarding medical collections contained enough information toinfer medical details about consumers, such as the type of treatment they had received.The ability to discern from a credit report that a consumer may have received treatmentfrom a neonatal clinic, a fertility clinic, a mental health provider, or an AIDS clinic hasserious implications for medical privacy, and could potentially facilitate discriminatorytreatment. While section 604 (g) of the Fair Credit Reporting Act prohibits furnishing of medical data in connection with employment, credit, or insurance transactions,consumers also complain that reporting collection accounts without identifying theoriginal creditor makes it difficult for consumers to decipher their own reports. It is theunderstanding of researchers that current market practices limit the level of detail in



43

reports provided to employers, aggregating information in such a way that individualcreditors are not identified, and an employer would be unlikely to be able to makespecific inferences about an applicant’s or employee’s medical condition. Nonetheless,the presence of this information among the data held at the repository level is troublingand deserving of further attention.



44

VIII. How to Improve the System

A. Require creditors to immediately provide to any consumer who experiences an

adverse action as a result of their credit reports or credit scores a copy of the

credit reports and scores used to arrive at that decision free of charge and permit disputes to be immediately resubmitted for reconsideration.

All consumers who experience an adverse action based on one or more credit reports orscores (such as having a loan or insurance application denied, being charged higher thanprime rates, or receiving less favorable terms) should immediately be given a copy of both the full report or reports used to derive that score and the related credit scoreswithout having to pay any additional fee. These reports should identify any entries thatare lowering the consumer’s score and indicate the impact (either the point valuededucted for that entry or the proportional impact of that entry relative to otherderogatory entries in the report). The consumer should then be allowed to identify any

errors or out of date information, provide documentation, and be reevaluated for primerates.

The additional cost to lenders and businesses of providing these reports immediatelywould be minimal. Since they already posses the report in paper or electronic form, theywould merely have to copy or print this report.

Simply providing consumers with the name and contact information of the consumerreporting agency or agencies that provided the information used to arrive at the decisionis insufficient because it creates an unnecessary obstacle and, especially for non-mortgage applications, the report a consumer will receive after submitting a request mayvery likely differ from the report the creditor reviewed. Errors from duplicate scoresand/or mixed reports that may result from incomplete or incorrect keying of informationduring the file request will not be apparent if the consumer correctly requests his or herfile. One in ten consumer applications results in an additional report being returned bythe repository.

B. Require decisions based on a single repository’s credit report or credit score

that result in anything less than the most favorable pricing to immediately trigger

a re-evaluation based on all three repositories at no additional cost.

Lenders and other credit data users have a desire to keep their underwriting costs low.This is a legitimate desire so long as consumers are not harmed in the process. Somereduce costs by underwriting certain decisions with only one credit report. For example,a lender may offer pre-approval letters based on only one report, or underwrite homeequity lines of credit or second mortgages with a single report. Given the wide rangebetween scores for a typical consumer and the frequency with which major accounts areomitted from credit reports, such practices have serious negative implications forconsumers.



45

Measures should be put in place to protect consumers from any negative impact resultingfrom such underwriting practices. A simple solution would be to require all decisionsbased on credit to use information from all three repositories. However, this could resultin higher costs and reduced availability of products such as pre-approval letters that arebeneficial to consumers.

Alternatively, lenders and other credit data users could be permitted to continueunderwriting based on one report, so long as any adverse impact based on informationfrom a single repository immediately triggers a re-evaluation with information from allthree repositories at no additional cost to the consumer. In this manner, businesses couldcontinue to save on underwriting costs for consumers with very good credit, butconsumers with less than perfect credit would not be forced to continue to pay a highprice for inaccuracies, inconsistencies, or incompleteness on any one credit report.

C. Strengthen requirements for complete and accurate reporting of account

information to credit repositories, and maintenance of consumer data by the

repositories, with adequate oversight and penalties for non-compliance.

Many errors in credit reports can be attributed to the practices of creditors and othercredit data users rather than to repositories. For example, some data furnishers may notreport to every credit bureau. Others may consciously misreport or omit informationregarding an account in order to prevent other lenders from approaching a valuablecustomer with competing offers (such as credit card lenders not reporting the trueavailable credit amount so that the borrower appears to have a much higher debt-to-available credit ratio and appears to pose greater risk when other lenders review the creditreport). Appropriate government entities such as the Federal Trade Commission andfederal banking regulators should require accurate and complete reporting of creditinformation to the repositories by any entity that uses credit data to make evaluations andconduct regular examinations for compliance. In addition to scrutinizing reportingentities, a government entity (such as the Federal Trade Commission) should audit therepositories’ records on a regular basis to identify data furnishers who report incompleteor incorrect information to the repositories. Such activity should be subject to fines orother penalties for non-compliance. These audits should also assess the overall accuracyof data maintained by the credit repositories, with appropriate fines or other penalties forinaccuracy.

Some may perceive tension between consumers’ interest in keeping their informationprivate and their interest in having evaluations of their creditworthiness be based on anaccurate record of their past behavior. However, consumers generally object toinformation sharing for secondary purposes, not in the regulated Fair Credit ReportingAct context, provided it is subject to Fair Information Practices. The cost of incorrectinformation is high, and it is possible to simultaneously serve both consumer interestsreasonably well.

Not all providers of consumer services use credit records or credit scores to determineconsumer eligibility, or pricing. However, those that do should be required to complete



46

the cycle of information and report complete and accurate information back to the creditrepositories. Information about any account that was underwritten with a report from oneor more credit repositories should be reported to those repositories as frequently as theconsumer is obligated to make payments. Collection agencies should be required toreport on the status of collections at least once every six months.

D. Establish meaningful oversight of the development of credit scoring systems.

Despite the fact that consumer access to, and pricing for, vital services such asmortgages, general consumer credit, insurance, rental housing, and utilities isincreasingly dictated by the automated evaluation of credit, there is no governmentoversight of the design of these systems. The calculations behind credit scores, a fact of life for the American consumer, remain shrouded in secrecy.

The design of credit scoring systems involves a number of deliberate choices that canhave a dramatic impact on consumers and can result in systems that are flawed or unfair.

These choices can range from determining the relative impact of various consumeractions to establishing the system defaults for cases where information such as date of last activity is not reported, to designing the logic for interpreting public records orcontradictory information reported for an account.

A wide variety of entities have developed scoring models30, including Fair Isaac andCompany, large mortgage lenders (such as Countrywide and GE Capital), the FederalHousing Administration and Department of Veterans Affairs loan guarantee programs,the Government Sponsored Enterprises (GSEs) Fannie Mae and Freddie Mac, privatemortgage insurance companies (such as PMI Mortgage Insurance Company andMortgage Guarantee Insurance Corporation), and insurance companies. However, theonly federal review of the fairness of any such models was a HUD review of the GSEsystems conducted in 2000, the findings of which are expected to be released soon31.While the delayed release will limit the relevance of this review because the GSEs havemade significant changes to their automated underwriting systems since the review wasconducted, we recommend other agencies follow this example and conduct full reviewsof all scoring systems in the marketplace.

We recommend that appropriate government agencies, such as HUD, the Federal TradeCommission, and state insurance departments conduct regular, comprehensiveevaluations of the validity and fairness of all credit scoring systems, including anyautomated mortgage underwriting systems, insurance underwriting systems, tenant andemployee screening systems, or any other systems or software that uses credit data as partof its evaluation of consumers, and report to Congress with its findings. Theseevaluations should be conducted and released in a timely fashion so that developers canreact to any recommendations and so the reviews do not become outdated as newversions of scoring software are developed and distributed. Strong oversight of scoring

30 Straka, John. 2000. A Shift in the Mortgage Landscape: the 1990s Move to Automated CreditEvaluations. Journal of Housing Research. Volume 11, Issue 2.31 Felsenthal, Mark. “HUD Secretary – mortgage software bias study out soon.” Reuters. October 22, 2002.



47

systems that identifies and protects consumers from any abuses will foster consumerconfidence in these powerful and increasingly utilized evaluation tools.

E. Address important questions and conduct further research.

In the course of conducting this study, several questions arose which are notcomprehensively addressed in this report, but are deserving of further attention andresearch. This report primarily addresses the impact of wide variations in credit scoresand credit data on consumers who are seeking credit – particularly mortgages. Futurestudies should explore the impact of these variations on insurance availability andaffordability, given the recent, dramatic increase in the use of credit scores as aninsurance underwriting tool. In addition, further research should address the impact of data and credit score variations on consumers as a result of other applications, such astenant screening and employee screening. Additional research could assess the value toconsumers of fee-based credit monitoring services.

Other topics raised in this report, but not exhaustively addressed, include determining thevalue to consumers of credit re-scoring relative to other means of credit data validation,the impact of anti-competitive market forces surrounding credit re-scoring, the privacyconcerns surrounding the appearance of medical related information in credit reports, andways to protect consumers from abusive applications of such medical information. TheFTC should promptly develop and require a mechanism to obscure medical debtor namesin credit reports.

The Fair Credit Reporting Act prohibits states from enacting any laws that provideprotections beyond those guaranteed by federal statute. On January 1, 2004 thisprovision will expire, although the federal law will otherwise remain in place. Contraryto some characterizations, the entire act will not “sunset” on this date. This prohibitionon supplemental state protections should not be extended, and if any changes to the FairCredit Reporting Act are to be made at the federal level, they should result in greaterconsumer protections and address the problems raised in this and other research.



IX. Recommendations for Consumers

Many of the concerns raised by this study address structural issues regarding the systemof reporting and evaluating credit, which are beyond the scope of most consumers toinfluence. However, there are some steps consumers can take to reduce the likelihood of

errors occurring, or to address them when they arise.

? Maintain consistency in credit applications: use your full legal name when applyingfor credit. If you have a generational title (Sr., Jr., III) always specify this.

? Review your credit record regularly by purchasing a credit report and score from eachmajor credit repository once a year. The repositories can be contacted at the followingphone numbers and website addresses: Equifax (800) 685-1111 or www.equifax.com;Experian (888) EXPERIAN or www.experian.com; Trans Union (800) 888-4213 orwww.transunion.com.

? Prior to applying for a mortgage, consider obtaining a current copy of your creditreport and score from each major repository, and review it for errors.

? Dispute any errors that appear on your credit report by contacting the creditrepository. However, avoid “credit repair” businesses that claim to be able to erasevalid items in consumers’ credit histories.

? Don’t underrate your credit. Ask for specifics if a lender tells you that you have badcredit and don’t qualify. Currently lenders do not have to tell you the specifics, orshow you the credit report that they review, but they are permitted to. If a lenderrefuses to talk to you about the specifics of your credit report, consider a different

lender.

? If you have complaints about your credit report and are unable to have them quicklyresolved, contact the Federal Trade Commission at 1-877-FTC-HELP orwww.ftc.gov.

US Federal Reserve: 121702CFA NCRA Credit Score Report Final

Documents

Transcript of US Federal Reserve: 121702CFA NCRA Credit Score Report Final